PyPI · #1459 most downloaded on PyPI

openhands-tools

1.49.29.8M downloads/moOpenHands/software-agent-sdk ↗New releases in your daily digest, changelog included.

OpenHands Tools - Runtime tools for AI agents

Last release today
17 Sep 2026
Ships on a steady schedule
a new release about every 1 weeks
Nearly every release is documented
notes for 60 of the last 60 stable releases
Nothing withdrawn
no release was ever pulled
11 months old
94 releases · first in 2025

Release timeline

94 releases since 2025

20252026

One column per month.

Releases

1.12.05 Mar 2026Security
Update orjson to 3.11.7 to address CVE-2025-67221 by @aivong-openhands in https://github.com/OpenHands/software-agent-sdk/pull/2268
ACP / Agent Client Protocol landed end-to-end: new ACPAgent, visualization of ACP tool calls, and ask_agent support via fork_session.

Agent-server got more capable and safer: new endpoints for LLM models/providers, better WebSocket auth (headers), improved event search behavior, secrets handling fixes, and a Bedrock safety fix (don’t forward api_key).

Subagents + skills got a big upgrade: file-based agent definitions (Markdown + YAML frontmatter), built-in specialized agent types (Explore/Bash), support for skills/LLM profiles/max-iterations, and new install utilities for AgentSkills.

CI and quality gates tightened: stronger API breakage detection/reporting (incl. oasdiff + release gating + deprecation policy), PR review agent guardrails, and a code review rule to block version bumps.

DevEx + platform compatibility improvements: multiple terminal/tmux reliability fixes (incl. macOS heredoc hang + tmux socket isolation), Windows import guards, plus dependency/model config updates (e.g., Qwen, Gemini preview, GPT-* codex/variant detection, security CVE bumps).

What's Changed

feat: Add ACPAgent for Agent Client Protocol integration by @simonrosenberg in https://github.com/OpenHands/software-agent-sdk/pull/2133

chore: add version bump blocking rule to code review guide by @xingyaoww in https://github.com/OpenHands/software-agent-sdk/pull/2161

Release v1.11.5 by @all-hands-bot in https://github.com/OpenHands/software-agent-sdk/pull/2160

Fix API breakage check robustness and reporting by @enyst in https://github.com/OpenHands/software-agent-sdk/pull/2098

ci: skip api breakage check when prev lacks all by @enyst in https://github.com/OpenHands/software-agent-sdk/pull/2165

Add ask_agent support to ACPAgent via fork_session by @simonrosenberg in https://github.com/OpenHands/software-agent-sdk/pull/2145

feat(sdk): Add TestLLM class for testing without mocking LiteLLM by @VascoSch92 in https://github.com/OpenHands/software-agent-sdk/pull/2016

DRAFT: docs(AGENTS): how to reply to GitHub inline review threads via REST API by @enyst in https://github.com/OpenHands/software-agent-sdk/pull/2090

fix(terminal): avoid sending C-l in TmuxTerminal.clear_screen() by @rbren in https://github.com/OpenHands/software-agent-sdk/pull/2166

fix(terminal): use dedicated tmux socket to isolate from user sessions by @rbren in https://github.com/OpenHands/software-agent-sdk/pull/2167

fix(Makefile): help and pre-commit by @VascoSch92 in https://github.com/OpenHands/software-agent-sdk/pull/2168

context: load project skills from git root by @enyst in https://github.com/OpenHands/software-agent-sdk/pull/2164

fix: preserve conversation updated_at across server restarts by @rbren in https://github.com/OpenHands/software-agent-sdk/pull/2172

ci: Gate API breakage failures on release PRs by @enyst in https://github.com/OpenHands/software-agent-sdk/pull/2173

Visualize ACP tool calls by @simonrosenberg in https://github.com/OpenHands/software-agent-sdk/pull/2162

Fix long heredoc commands hanging in SubprocessTerminal on macOS by @jpshackelford in https://github.com/OpenHands/software-agent-sdk/pull/2182

fix: Include boto3 extra in agent-server Docker image by @rbren in https://github.com/OpenHands/software-agent-sdk/pull/2188

Add LLM models and providers endpoints to agent-server by @rbren in https://github.com/OpenHands/software-agent-sdk/pull/2187

Do not forward api_key to Bedrock calls by @enyst in https://github.com/OpenHands/software-agent-sdk/pull/2195

feat(tools): task tool set by @VascoSch92 in https://github.com/OpenHands/software-agent-sdk/pull/2143

fix(tools): remove BrowserToolSet from package init to reduce downstream bundle size by @malhotra5 in https://github.com/OpenHands/software-agent-sdk/pull/2197

feat(delegate): File-based agent definitions with markdown + YAML frontmatter by @VascoSch92 in https://github.com/OpenHands/software-agent-sdk/pull/2183

agent-server: support WebSocket auth via headers by @enyst in https://github.com/OpenHands/software-agent-sdk/pull/1814

Add dashscope/qwen3.5-flash-2026-02-23 model configuration by @juanmichelini in https://github.com/OpenHands/software-agent-sdk/pull/2207

chore(deps): bump python-multipart from 0.0.20 to 0.0.22 by @dependabot[bot] in https://github.com/OpenHands/software-agent-sdk/pull/2210

chore(deps): bump werkzeug from 3.1.1 to 3.1.6 by @dependabot[bot] in https://github.com/OpenHands/software-agent-sdk/pull/2214

chore(deps): bump cryptography from 46.0.3 to 46.0.5 by @dependabot[bot] in https://github.com/OpenHands/software-agent-sdk/pull/2211

chore(deps): bump pyasn1 from 0.6.1 to 0.6.2 by @dependabot[bot] in https://github.com/OpenHands/software-agent-sdk/pull/2212

Filter public skills by default marketplace by @neubig in https://github.com/OpenHands/software-agent-sdk/pull/2205

chore(deps): bump pypdf from 6.1.1 to 6.7.2 by @dependabot[bot] in https://github.com/OpenHands/software-agent-sdk/pull/2213

nit(tools): small refactoring test files by @VascoSch92 in https://github.com/OpenHands/software-agent-sdk/pull/2215

chore(deps): bump authlib from 1.6.5 to 1.6.6 by @dependabot[bot] in https://github.com/OpenHands/software-agent-sdk/pull/2219

chore(deps): bump virtualenv from 20.34.0 to 20.36.1 by @dependabot[bot] in https://github.com/OpenHands/software-agent-sdk/pull/2220

ci: enforce agent-server REST API deprecation policy by @enyst in https://github.com/OpenHands/software-agent-sdk/pull/2232

fix(sdk): handle newlines in JSON-stringified dict arguments by @VascoSch92 in https://github.com/OpenHands/software-agent-sdk/pull/2217

fix(tools): merge subagents metrics (TaskToolSet) by @VascoSch92 in https://github.com/OpenHands/software-agent-sdk/pull/2222

fix(tools): merge subagents metrics (DelegateTool) by @VascoSch92 in https://github.com/OpenHands/software-agent-sdk/pull/2221

fix(tool): race condition in dynamic Action wrapper class creation by @VascoSch92 in https://github.com/OpenHands/software-agent-sdk/pull/2224

Fix GLM-5 preflight check by filtering SDK-specific parameters by @juanmichelini in https://github.com/OpenHands/software-agent-sdk/pull/2194

docs(subagent): document loader invariants for file-based agents by @enyst in https://github.com/OpenHands/software-agent-sdk/pull/2231

refactoring(sdk): Remove hardcoded header from get_factory_info() by @VascoSch92 in https://github.com/OpenHands/software-agent-sdk/pull/2234

fix(sdk): add gpt-5.2-codex, gpt-5.3-codex, and gpt-5.2 to model-variant detection by @enyst in https://github.com/OpenHands/software-agent-sdk/pull/2238

ci: use oasdiff for agent-server REST API breakage detection by @enyst in https://github.com/OpenHands/software-agent-sdk/pull/2240

fix: override server_image default to None in DockerDevWorkspace by @simonrosenberg in https://github.com/OpenHands/software-agent-sdk/pull/2243

PR review agent: avoid approving eval-risk behavior changes by @enyst in https://github.com/OpenHands/software-agent-sdk/pull/2246

Add server-base-path support for VSCode in path-based routing by @chuckbutkus in https://github.com/OpenHands/software-agent-sdk/pull/2241

feat: autotitle conversations on first user message by @rbren in https://github.com/OpenHands/software-agent-sdk/pull/2225

PR review agent: make eval-risk approval policy repo-specific by @enyst in https://github.com/OpenHands/software-agent-sdk/pull/2254

fix: Include secrets in system prompt when added via update_secrets() by @rbren in https://github.com/OpenHands/software-agent-sdk/pull/2171

Fix LiteLLM cost tracking for provider-prefixed models by @ShreySatapara in https://github.com/OpenHands/software-agent-sdk/pull/2257

Set default temperature to None instead of 0.0 by @neubig in https://github.com/OpenHands/software-agent-sdk/pull/1989

fix(tools): guard Unix-only imports in terminal package for Windows by @WolffM in https://github.com/OpenHands/software-agent-sdk/pull/2096

feat(delegate): Built-in specialized agent types (Explore, Bash) by @VascoSch92 in https://github.com/OpenHands/software-agent-sdk/pull/2201

Change eval_limit from choice to free-form string input by @simonrosenberg in https://github.com/OpenHands/software-agent-sdk/pull/2261

fix: cap max_output_tokens when using max_tokens fallback by @csmith49 in https://github.com/OpenHands/software-agent-sdk/pull/2264

docs: README for condenser module by @csmith49 in https://github.com/OpenHands/software-agent-sdk/pull/2262

fix: use query parameters for git API endpoints to preserve path slashes by @chuckbutkus in https://github.com/OpenHands/software-agent-sdk/pull/2249

chore(deps): bump mcp from 1.17.0 to 1.23.0 by @dependabot[bot] in https://github.com/OpenHands/software-agent-sdk/pull/2226

chore(deps): bump pypdf from 6.7.2 to 6.7.4 by @dependabot[bot] in https://github.com/OpenHands/software-agent-sdk/pull/2265

chore(deps): bump pypdf from 6.7.4 to 6.7.5 by @dependabot[bot] in https://github.com/OpenHands/software-agent-sdk/pull/2267

Add gemini-3.1-pro-preview model to reasoning models #2269 by @mayeco in https://github.com/OpenHands/software-agent-sdk/pull/2270

Show full dynamic context in SystemPromptEvent.visualize by @xingyaoww in https://github.com/OpenHands/software-agent-sdk/pull/2273

fix(sdk/subagent): fix get_factory_info() by @VascoSch92 in https://github.com/OpenHands/software-agent-sdk/pull/2271

Update orjson to 3.11.7 to address CVE-2025-67221 by @aivong-openhands in https://github.com/OpenHands/software-agent-sdk/pull/2268

Fix: Security analyzer ignores LLM security_risk when no analyzer is configured by @juanmichelini in https://github.com/OpenHands/software-agent-sdk/pull/2130

chore(deps): bump fastmcp from 2.12.4 to 2.14.0 by @dependabot[bot] in https://github.com/OpenHands/software-agent-sdk/pull/2266

Add AGENTS.md for model addition guidance in .github/run-eval by @juanmichelini in https://github.com/OpenHands/software-agent-sdk/pull/2284

Add explicit rules to prevent modifying existing models in AGENTS.md by @juanmichelini in https://github.com/OpenHands/software-agent-sdk/pull/2288

Split AGENTS.md: Move model addition guide to ADDINGMODEL.md by @juanmichelini in https://github.com/OpenHands/software-agent-sdk/pull/2291

add ability to run chrome as root by @rbren in https://github.com/OpenHands/software-agent-sdk/pull/2169

Add gpt-5-3-codex model to resolve_model_config.py by @juanmichelini in https://github.com/OpenHands/software-agent-sdk/pull/2292

Insist on integration tests on the model addition AGENTS.MD by @juanmichelini in https://github.com/OpenHands/software-agent-sdk/pull/2293

feat(sdk/subagent): Support skills in file-based Agents by @VascoSch92 in https://github.com/OpenHands/software-agent-sdk/pull/2260

feat(sdk/subagent): Support max_iteration_per_run in file-based agent definitions by @VascoSch92 in https://github.com/OpenHands/software-agent-sdk/pull/2263

feat(plugin): Add installed plugins management utilities by @xingyaoww in https://github.com/OpenHands/software-agent-sdk/pull/2031

Fix duplicate observations for same tool_call_id after crash recovery by @csmith49 in https://github.com/OpenHands/software-agent-sdk/pull/2300

Fix API breakage workflow exit codes by @enyst in https://github.com/OpenHands/software-agent-sdk/pull/2309

Fix prompt_cache_retention for Azure by @enyst in https://github.com/OpenHands/software-agent-sdk/pull/2311

Disable prompt_cache_retention for Azure models by @enyst in https://github.com/OpenHands/software-agent-sdk/pull/2312

fix(examples): example no. 41 by @VascoSch92 in https://github.com/OpenHands/software-agent-sdk/pull/2314

nit(examples): example no. 40 by @VascoSch92 in https://github.com/OpenHands/software-agent-sdk/pull/2315

fix(examples): example no. 25 by @VascoSch92 in https://github.com/OpenHands/software-agent-sdk/pull/2313

fix(security): Upgrade fastmcp to v3 to address CVE-2025-69872 by @aivong-openhands in https://github.com/OpenHands/software-agent-sdk/pull/2294

Fix: Add 'gpt5' to tool-preset argparse choices by @juanmichelini in https://github.com/OpenHands/software-agent-sdk/pull/2306

feat(sdk/subagent): Support LLM profiles in subagents by @VascoSch92 in https://github.com/OpenHands/software-agent-sdk/pull/2258

Normalize OpenAPI schema for oasdiff by @enyst in https://github.com/OpenHands/software-agent-sdk/pull/2318

chore(deps): bump authlib from 1.6.6 to 1.6.7 by @dependabot[bot] in https://github.com/OpenHands/software-agent-sdk/pull/2321

fix: events search API blocks during agent step (FIFOLock contention) by @rbren in https://github.com/OpenHands/software-agent-sdk/pull/2296

fix(examples): example no.38 by @VascoSch92 in https://github.com/OpenHands/software-agent-sdk/pull/2316

feat: add AgentSkills install utilities by @enyst in https://github.com/OpenHands/software-agent-sdk/pull/2320

fix: set delete_on_close default to True by @xingyaoww in https://github.com/OpenHands/software-agent-sdk/pull/2328

chore: add design-principles skill linking to SDK arch docs by @xingyaoww in https://github.com/OpenHands/software-agent-sdk/pull/2329

Route ACP agents through LiteLLM proxy in CI by @simonrosenberg in https://github.com/OpenHands/software-agent-sdk/pull/2326

New Contributors

@chuckbutkus made their first contribution in https://github.com/OpenHands/software-agent-sdk/pull/2241

@ShreySatapara made their first contribution in https://github.com/OpenHands/software-agent-sdk/pull/2257

@WolffM made their first contribution in https://github.com/OpenHands/software-agent-sdk/pull/2096

@mayeco made their first contribution in https://github.com/OpenHands/software-agent-sdk/pull/2270

@aivong-openhands made their first contribution in https://github.com/OpenHands/software-agent-sdk/pull/2268

Full Changelog: https://github.com/OpenHands/software-agent-sdk/compare/v1.11.5...v1.12.0

1.11.520 Feb 2026Deprecated

Handle deprecated enable_truncation field in TextContent for backward compatibility by @xingyaoww in https://github.com/OpenHands/software-agent-sdk/p…

Add support for Sonnet 4.6, GLM 5, Gemini 3.1 Pro, Minimax M2.5
Multiple reliability fixes

What's Changed

Release v1.11.4 by @all-hands-bot in https://github.com/OpenHands/software-agent-sdk/pull/2011
Fix workflow_run trigger for version-bump-prs by @xingyaoww in https://github.com/OpenHands/software-agent-sdk/pull/2014
Remove enterprise/pyproject.toml updates from version bump workflow by @xingyaoww in https://github.com/OpenHands/software-agent-sdk/pull/2015
chore: move skills directory to .agents by @enyst in https://github.com/OpenHands/software-agent-sdk/pull/1970
feat: Load project skills (AGENTS.md, etc.) in PR review action by @xingyaoww in https://github.com/OpenHands/software-agent-sdk/pull/2017
Feat: session recording agent's browser sessions by @malhotra5 in https://github.com/OpenHands/software-agent-sdk/pull/1731
feat: Add current_datetime field to AgentContext by @xingyaoww in https://github.com/OpenHands/software-agent-sdk/pull/2012
docs: trim blustery wording in docs update prompt by @smolpaws in https://github.com/OpenHands/software-agent-sdk/pull/2019
fix(ci): block PR review auto-run for author_association NONE by @enyst in https://github.com/OpenHands/software-agent-sdk/pull/2020
fix(pr-review): avoid blank description in prompt by @enyst in https://github.com/OpenHands/software-agent-sdk/pull/2021
feat(security): add GraySwan Cygnal security analyzer by @neubig in https://github.com/OpenHands/software-agent-sdk/pull/1952
fix: add circular reference detection to _process_schema_node by @neubig in https://github.com/OpenHands/software-agent-sdk/pull/1956
docs(review): clarify approval guidance by @enyst in https://github.com/OpenHands/software-agent-sdk/pull/2022
fix: bundle browser recording JS files in PyInstaller build by @malhotra5 in https://github.com/OpenHands/software-agent-sdk/pull/2025
Fix issue with remote docker workspaces. by @hamiltop in https://github.com/OpenHands/software-agent-sdk/pull/1807
Add GLM-5 to expected models by @juanmichelini in https://github.com/OpenHands/software-agent-sdk/pull/2029
fix(llm): ensure second LLM gets independent metrics and telemetry by @xingyaoww in https://github.com/OpenHands/software-agent-sdk/pull/1997
fix(pr-review): Set trace metadata within active span context by @neubig in https://github.com/OpenHands/software-agent-sdk/pull/2062
Add pr_url metadata to Laminar logging in PR review bot by @neubig in https://github.com/OpenHands/software-agent-sdk/pull/2068
Handle deprecated enable_truncation field in TextContent for backward compatibility by @xingyaoww in https://github.com/OpenHands/software-agent-sdk/pull/2027
fix(ci): Fix PR review evaluation workflow artifact detection by @neubig in https://github.com/OpenHands/software-agent-sdk/pull/2070
DRAFT: Add feature-release-rollout skill for multi-repo feature propagation by @neubig in https://github.com/OpenHands/software-agent-sdk/pull/2069
ci: API breakage checks for SDK (Griffe) by @enyst in https://github.com/OpenHands/software-agent-sdk/pull/1098
ci: require deprecation marker before exported symbol removal by @enyst in https://github.com/OpenHands/software-agent-sdk/pull/2072
ci: extend API breakage checks to openhands-workspace by @enyst in https://github.com/OpenHands/software-agent-sdk/pull/2075
Update issue number in integration runner workflow by @enyst in https://github.com/OpenHands/software-agent-sdk/pull/2079
SDK: bound latest-user-message scan in Agent.step by @enyst in https://github.com/OpenHands/software-agent-sdk/pull/1844
ci: extend API breakage checks to openhands-tools by @enyst in https://github.com/OpenHands/software-agent-sdk/pull/2080
docs(sdk): add scoped AGENTS.md by @enyst in https://github.com/OpenHands/software-agent-sdk/pull/2081
docs: add package-scoped AGENTS.md guides by @enyst in https://github.com/OpenHands/software-agent-sdk/pull/2086
docs: align AGENTS.md guidance across repo by @enyst in https://github.com/OpenHands/software-agent-sdk/pull/2087
Enforce deprecation period for removed public methods by @enyst in https://github.com/OpenHands/software-agent-sdk/pull/2083
fix: validate provider prefixes in unverified model list by @enyst in https://github.com/OpenHands/software-agent-sdk/pull/1668
feat(pr-review): Support multiple models with random selection for A/B testing by @neubig in https://github.com/OpenHands/software-agent-sdk/pull/2024
feat(skills): load user skills from ~/.agents/skills by @cbagwell in https://github.com/OpenHands/software-agent-sdk/pull/2091
refactor: update PUBLIC_SKILLS_REPO to OpenHands/extensions by @neubig in https://github.com/OpenHands/software-agent-sdk/pull/2085
security: fix HTTP-related CVEs in transitive dependencies by @raymyers in https://github.com/OpenHands/software-agent-sdk/pull/2094
security: fix protobuf and pillow CVEs in transitive dependencies by @raymyers in https://github.com/OpenHands/software-agent-sdk/pull/2095
feat(sdk): add fallback strategy (LLM) by @VascoSch92 in https://github.com/OpenHands/software-agent-sdk/pull/2093
Update sdk_ref default in run-eval.yml during release preparation by @juanmichelini in https://github.com/OpenHands/software-agent-sdk/pull/1938
Add claude-sonnet-4-6 to expected models by @juanmichelini in https://github.com/OpenHands/software-agent-sdk/pull/2102
fix: make skill loading resilient to individual skill errors by @jpelletier1 in https://github.com/OpenHands/software-agent-sdk/pull/2108
Remove jade-spark-2862 alias, add its settings to minimax-m2.5 by @neubig in https://github.com/OpenHands/software-agent-sdk/pull/2106
fix: Forcing minimum condenser progress by @csmith49 in https://github.com/OpenHands/software-agent-sdk/pull/2107
fix: disable vision for GLM-5 model by @juanmichelini in https://github.com/OpenHands/software-agent-sdk/pull/2111
Add preflight LLM check before dispatching evaluations by @neubig in https://github.com/OpenHands/software-agent-sdk/pull/2109
Add configurable file editing toolset support by @neubig in https://github.com/OpenHands/software-agent-sdk/pull/2077
chore: Rename code-review.md to custom-codereview-guide.md by @xingyaoww in https://github.com/OpenHands/software-agent-sdk/pull/2121
Disable uv Actions cache in PR review agent workflow by @enyst in https://github.com/OpenHands/software-agent-sdk/pull/2123
fix: Move litellm install before model loading and use correct API key by @juanmichelini in https://github.com/OpenHands/software-agent-sdk/pull/2118
Fix: Make litellm import lazy in resolve_model_config.py by @juanmichelini in https://github.com/OpenHands/software-agent-sdk/pull/2125
feat(hooks): add async hook execution support by @ixchio in https://github.com/OpenHands/software-agent-sdk/pull/1849
feat(condenser): Explicit view properties by @csmith49 in https://github.com/OpenHands/software-agent-sdk/pull/2116
Fix: Add 180-minute timeout to integration tests to prevent indefinite hangs by @juanmichelini in https://github.com/OpenHands/software-agent-sdk/pull/2131
Fix: Add claude-sonnet-4-6 to EXTENDED_THINKING_MODELS by @juanmichelini in https://github.com/OpenHands/software-agent-sdk/pull/2138
Enable Datadog persistence by default in eval job by @juanmichelini in https://github.com/OpenHands/software-agent-sdk/pull/2140
Fix PR review workflow by @enyst in https://github.com/OpenHands/software-agent-sdk/pull/2126
fix: Disable browser tools in integration tests to fix ProcessPoolExecutor hang by @neubig in https://github.com/OpenHands/software-agent-sdk/pull/2149
feat: Add Gemini 3.1 Pro to evaluation model config by @neubig in https://github.com/OpenHands/software-agent-sdk/pull/2153
CI: simplify pr-review gating and skip forks by @enyst in https://github.com/OpenHands/software-agent-sdk/pull/2159
Add API compliance test framework for malformed message patterns by @csmith49 in https://github.com/OpenHands/software-agent-sdk/pull/2155
Update integration tests to use claude-sonnet-4-6 by @xingyaoww in https://github.com/OpenHands/software-agent-sdk/pull/2113
Docs: fix README links and typos by @enyst in https://github.com/OpenHands/software-agent-sdk/pull/2158
Add claude-sonnet-4-6 to verified models by @juanmichelini in https://github.com/OpenHands/software-agent-sdk/pull/2104
Add event-sourcing system benchmarks by @simonrosenberg in https://github.com/OpenHands/software-agent-sdk/pull/2032

New Contributors

@smolpaws made their first contribution in https://github.com/OpenHands/software-agent-sdk/pull/2019
@jpelletier1 made their first contribution in https://github.com/OpenHands/software-agent-sdk/pull/2108

Full Changelog: https://github.com/OpenHands/software-agent-sdk/compare/v1.11.4...v1.11.5

Open source →

1.11.411 Feb 2026

fix: serialize Laminar span_context UUIDs as strings for JSON compatibility by @neubig in https://github.com/OpenHands/software-agent-sdk/pull/2000

What's Changed

fix: serialize Laminar span_context UUIDs as strings for JSON compatibility by @neubig in https://github.com/OpenHands/software-agent-sdk/pull/2000
feat: add previous review context to PR review agent by @xingyaoww in https://github.com/OpenHands/software-agent-sdk/pull/1991
Separate version bump PRs into standalone workflow by @xingyaoww in https://github.com/OpenHands/software-agent-sdk/pull/2003
Release v1.11.3 by @all-hands-bot in https://github.com/OpenHands/software-agent-sdk/pull/2002
fix: fetch latest reviews/threads instead of oldest by @xingyaoww in https://github.com/OpenHands/software-agent-sdk/pull/2004
Fix poetry lock command in version-bump-prs workflow by @xingyaoww in https://github.com/OpenHands/software-agent-sdk/pull/2006
Remove redundant poetry lock calls in version-bump-prs workflow by @xingyaoww in https://github.com/OpenHands/software-agent-sdk/pull/2008
Add MiniMax-M2.5 model configuration by @neubig in https://github.com/OpenHands/software-agent-sdk/pull/2007
Add pre-commit step to version-bump-prs workflow by @xingyaoww in https://github.com/OpenHands/software-agent-sdk/pull/2010

Full Changelog: https://github.com/OpenHands/software-agent-sdk/compare/v1.11.3...v1.11.4

Open source →

1.11.311 Feb 2026

Release v1.11.2 by @all-hands-bot in https://github.com/OpenHands/software-agent-sdk/pull/1976

What's Changed

Release v1.11.2 by @all-hands-bot in https://github.com/OpenHands/software-agent-sdk/pull/1976
fix: exclude reference markdown files in agentskills folders from being loaded as skills by @xingyaoww in https://github.com/OpenHands/software-agent-sdk/pull/1982
Add stop hook for pre-commit, pytest, and CI validation and add /api/hooks to agent-server by @xingyaoww in https://github.com/OpenHands/software-agent-sdk/pull/1878
feat(eval): add jade-spark-2862 model for evaluation by @neubig in https://github.com/OpenHands/software-agent-sdk/pull/1984
ci: switch integration tests to use eval proxy by @neubig in https://github.com/OpenHands/software-agent-sdk/pull/1985
fix: separate static system prompt from dynamic context for cross-conversation caching by @neubig in https://github.com/OpenHands/software-agent-sdk/pull/1890
Fix Laminar trace continuation for PR review evaluation by @neubig in https://github.com/OpenHands/software-agent-sdk/pull/1988
fix(llm): detect context window errors from Minimax via APIConnectionError by @neubig in https://github.com/OpenHands/software-agent-sdk/pull/1992
feat: add rejection_source field to UserRejectObservation by @xingyaoww in https://github.com/OpenHands/software-agent-sdk/pull/1995
docs(AGENTS.md): add GraphQL examples for resolving review threads by @neubig in https://github.com/OpenHands/software-agent-sdk/pull/1993

Full Changelog: https://github.com/OpenHands/software-agent-sdk/compare/v1.11.2...v1.11.3

Open source →

1.11.29 Feb 2026

Release v1.11.1 by @all-hands-bot in https://github.com/OpenHands/software-agent-sdk/pull/1921

What's Changed

Release v1.11.1 by @all-hands-bot in https://github.com/OpenHands/software-agent-sdk/pull/1921
Add litellm_proxy/gpt-oss-20b to resolve_model_config.py by @juanmichelini in https://github.com/OpenHands/software-agent-sdk/pull/1906
fix: Remove python-version from setup-uv in create-version-bump-prs job by @juanmichelini in https://github.com/OpenHands/software-agent-sdk/pull/1926
Add Claude Opus 4.6 model support by @neubig in https://github.com/OpenHands/software-agent-sdk/pull/1933
Set default sdk_ref to v1.11.1 in run-eval workflow by @juanmichelini in https://github.com/OpenHands/software-agent-sdk/pull/1935
feat(llm): add gpt-5.3-codex to subscription models by @enyst in https://github.com/OpenHands/software-agent-sdk/pull/1940
Add example for reconstructing OpenAI messages from events by @enyst in https://github.com/OpenHands/software-agent-sdk/pull/1916
feat: add /ready endpoint for proper Kubernetes readiness checks by @neubig in https://github.com/OpenHands/software-agent-sdk/pull/1810
fix(llm): Add Claude Opus 4.6 to model features for reasoning_effort support by @simonrosenberg in https://github.com/OpenHands/software-agent-sdk/pull/1941
nit(sdk): change usage_to_llm to return a MappingProxyType by @VascoSch92 in https://github.com/OpenHands/software-agent-sdk/pull/1930
fix: use Python 3.12 for openhands-cli version bump by @xingyaoww in https://github.com/OpenHands/software-agent-sdk/pull/1924
feat(ci): add model_ids and issue_number inputs to integration-runner by @neubig in https://github.com/OpenHands/software-agent-sdk/pull/1883
Add Laminar traces to PR review workflow by @juanmichelini in https://github.com/OpenHands/software-agent-sdk/pull/1949
fix(message): Responses tool image serialization (preserve file_editor images) by @Wangmerlyn in https://github.com/OpenHands/software-agent-sdk/pull/1895
fix: disable vision for glm-4.7 to prevent multimodal evaluation failures by @juanmichelini in https://github.com/OpenHands/software-agent-sdk/pull/1898
Fix tool serialization test flakiness by @enyst in https://github.com/OpenHands/software-agent-sdk/pull/1959
refactor: Use composite GitHub Action for PR review workflow by @xingyaoww in https://github.com/OpenHands/software-agent-sdk/pull/1927
chore: add review thread gate by @enyst in https://github.com/OpenHands/software-agent-sdk/pull/1962
fix: update LLM model and base URL for pr-review workflow by @xingyaoww in https://github.com/OpenHands/software-agent-sdk/pull/1968
feat: Add delayed evaluation for PR review with Laminar signals by @neubig in https://github.com/OpenHands/software-agent-sdk/pull/1954
Add context window size validation for local LLMs by @neubig in https://github.com/OpenHands/software-agent-sdk/pull/1961
feat(sdk): introduce LLMProfileStore for persisted LLM configurations by @VascoSch92 in https://github.com/OpenHands/software-agent-sdk/pull/1928
fix(pyinstaller): include delegate tool templates in bundle by @malhotra5 in https://github.com/OpenHands/software-agent-sdk/pull/1971
fix: handle PS1 metadata corruption in command output by @neubig in https://github.com/OpenHands/software-agent-sdk/pull/1817
Enhance planning agent to ask clarifying questions for ambiguous requests by @xingyaoww in https://github.com/OpenHands/software-agent-sdk/pull/1967
feat(examples): update critic example to demonstrate iterative refinement by @xingyaoww in https://github.com/OpenHands/software-agent-sdk/pull/1879

New Contributors

@VascoSch92 made their first contribution in https://github.com/OpenHands/software-agent-sdk/pull/1930

Full Changelog: https://github.com/OpenHands/software-agent-sdk/compare/v1.11.1...v1.11.2

Open source →

1.11.15 Feb 2026

Release v1.11.0 by @all-hands-bot in https://github.com/OpenHands/software-agent-sdk/pull/1884

What's Changed

Release v1.11.0 by @all-hands-bot in https://github.com/OpenHands/software-agent-sdk/pull/1884
Fix license by @enyst in https://github.com/OpenHands/software-agent-sdk/pull/1869
Add qwen3-coder-next to expected models by @juanmichelini in https://github.com/OpenHands/software-agent-sdk/pull/1892
utils(release): Fix poetry lock conflict and move openhands-cli step earlier by @xingyaoww in https://github.com/OpenHands/software-agent-sdk/pull/1888
Change qwen3-coder-next model from together.ai to openrouter by @juanmichelini in https://github.com/OpenHands/software-agent-sdk/pull/1900
Fix truncation at LLM message layer by @enyst in https://github.com/OpenHands/software-agent-sdk/pull/1838
Add qwen3-coder-30b-a3b-instruct model to resolve_model_config.py by @juanmichelini in https://github.com/OpenHands/software-agent-sdk/pull/1903
feat(llm): add gpt-5.2-codex to verified models by @juanmichelini in https://github.com/OpenHands/software-agent-sdk/pull/1893
fix: serialize tmux session creation to prevent race conditions by @neubig in https://github.com/OpenHands/software-agent-sdk/pull/1889
Update code-review skill with repo-specific approval guidelines by @xingyaoww in https://github.com/OpenHands/software-agent-sdk/pull/1904
fix(mcp): handle timeout errors gracefully in MCP tool execution by @neubig in https://github.com/OpenHands/software-agent-sdk/pull/1862
feat(llm): add Kimi K2.5 to verified models by @juanmichelini in https://github.com/OpenHands/software-agent-sdk/pull/1907
Add conversation.execute_tool() method for pre-run tool execution by @xingyaoww in https://github.com/OpenHands/software-agent-sdk/pull/1833
Clarify execute_tool bypasses confirmation/security checks by @enyst in https://github.com/OpenHands/software-agent-sdk/pull/1917
chore: remove blacksmith CI runners and use GitHub's default runners by @xingyaoww in https://github.com/OpenHands/software-agent-sdk/pull/1915
fix: revert agent-server Docker image to Python 3.12 by @neubig in https://github.com/OpenHands/software-agent-sdk/pull/1910
Cache Qwen3 tokenizer config for critic template tests by @enyst in https://github.com/OpenHands/software-agent-sdk/pull/1919
fix: wait for WebSocket terminal status to prevent event loss by @xingyaoww in https://github.com/OpenHands/software-agent-sdk/pull/1832
Add skill for debugging test-examples workflow by @xingyaoww in https://github.com/OpenHands/software-agent-sdk/pull/1887
feat(skills): support .agents/skills directory by @enyst in https://github.com/OpenHands/software-agent-sdk/pull/1914
fix: use blacksmith runners for all test jobs to fix coverage by @xingyaoww in https://github.com/OpenHands/software-agent-sdk/pull/1920

Full Changelog: https://github.com/OpenHands/software-agent-sdk/compare/v1.11.0...v1.11.1

Open source →

1.11.03 Feb 2026

Added support for login with LLM subscription (ChatGPT) - Docs: https://docs.openhands.dev/sdk/guides/llm-subscriptions

What's Changed

Simplify PR review prompt to use /github-pr-review skill by @xingyaoww in https://github.com/OpenHands/software-agent-sdk/pull/1792
Add gpt-5.2-codex model to evaluation config by @neubig in https://github.com/OpenHands/software-agent-sdk/pull/1835
Release v1.10.0 by @all-hands-bot in https://github.com/OpenHands/software-agent-sdk/pull/1827
Cap event history scanned by StuckDetector by @enyst in https://github.com/OpenHands/software-agent-sdk/pull/1829
ci: default to Python 3.13 (keep 3.12+ support) by @enyst in https://github.com/OpenHands/software-agent-sdk/pull/1836
Add integration-style tests for conversation restore behavior by @enyst in https://github.com/OpenHands/software-agent-sdk/pull/1799
Add run-eval skill for evaluation guidance by @neubig in https://github.com/OpenHands/software-agent-sdk/pull/1809
Add Kimi K2.5 and Qwen3 Max Thinking models to eval config by @neubig in https://github.com/OpenHands/software-agent-sdk/pull/1845
Redact cookie/token headers in LookupSecret serialization by @enyst in https://github.com/OpenHands/software-agent-sdk/pull/1822
Remove truncation for skill files (AGENTS.md, others) by @enyst in https://github.com/OpenHands/software-agent-sdk/pull/1842
Add GLM-4.7 to expected models by @juanmichelini in https://github.com/OpenHands/software-agent-sdk/pull/1854
Truncate terminal outputs before persisting events by @enyst in https://github.com/OpenHands/software-agent-sdk/pull/1823
fix(security): prevent auth token leakage in trajectory exports by @jpshackelford in https://github.com/OpenHands/software-agent-sdk/pull/1848
Avoid materializing full event history in Agent.init_state by @enyst in https://github.com/OpenHands/software-agent-sdk/pull/1840
feat(condenser): Hard context reset on unrecoverable error by @csmith49 in https://github.com/OpenHands/software-agent-sdk/pull/1596
Add guidance to avoid exposing raw secret values in conversation by @xingyaoww in https://github.com/OpenHands/software-agent-sdk/pull/1855
fix: prevent duplicate events in bash polling via order__gt filtering by @neubig in https://github.com/OpenHands/software-agent-sdk/pull/1816
fix: add MAINTAINERS file and update triage prompt to use it by @xingyaoww in https://github.com/OpenHands/software-agent-sdk/pull/1859
Add --check-browser flag for browser functionality testing by @raymyers in https://github.com/OpenHands/software-agent-sdk/pull/1329
feat: require semantic version name on eval github action by @neubig in https://github.com/OpenHands/software-agent-sdk/pull/1867
fix: Add AttributeError to escape_bash_special_chars exception handling by @daviskay75 in https://github.com/OpenHands/software-agent-sdk/pull/1871
feat(llm): Add subscription-based authentication for OpenAI Codex models by @xingyaoww in https://github.com/OpenHands/software-agent-sdk/pull/1682
fix: update enterprise/pyproject.toml before poetry lock in version bump workflow by @xingyaoww in https://github.com/OpenHands/software-agent-sdk/pull/1834
fix: several issues related to scalability by @Co1lin in https://github.com/OpenHands/software-agent-sdk/pull/1619
fix: unable to start a new v1 conversation when using the latest sdk code by @hieptl in https://github.com/OpenHands/software-agent-sdk/pull/1881
feat: store plan.md file in appropriate configuration folders by @hieptl in https://github.com/OpenHands/software-agent-sdk/pull/1876

New Contributors

@daviskay75 made their first contribution in https://github.com/OpenHands/software-agent-sdk/pull/1871

Full Changelog: https://github.com/OpenHands/software-agent-sdk/compare/v1.10.0...v1.11.0

Open source →

1.10.026 Jan 2026Deprecated

Deprecate safety_settings field in LLM configuration by @enyst in https://github.com/OpenHands/software-agent-sdk/pull/1803

What's Changed

feat(agent-server): Support plugin loading when starting conversations by @jpshackelford in https://github.com/OpenHands/software-agent-sdk/pull/1651
feat(plugin): Load commands as keyword-triggered skills by @jpshackelford in https://github.com/OpenHands/software-agent-sdk/pull/1676
feat(condenser): Multi-summary views by @csmith49 in https://github.com/OpenHands/software-agent-sdk/pull/1721
refactor: enforce tool immutability on conversation restore by @enyst in https://github.com/OpenHands/software-agent-sdk/pull/1788
feat(message): require explicit values for serialization control fields by @neubig in https://github.com/OpenHands/software-agent-sdk/pull/1798
Add NVIDIA Nemotron 3 Nano 30B model configuration by @juanmichelini in https://github.com/OpenHands/software-agent-sdk/pull/1804
Fix MCP HTTP session persistence across tool calls by @neubig in https://github.com/OpenHands/software-agent-sdk/pull/1740
refactor(message): move serialization control to to_chat_dict() parameters by @neubig in https://github.com/OpenHands/software-agent-sdk/pull/1800
Deprecate safety_settings field in LLM configuration by @enyst in https://github.com/OpenHands/software-agent-sdk/pull/1803
Hide pyinstaller vendored libraries from subprocesses by @cbagwell in https://github.com/OpenHands/software-agent-sdk/pull/1758
fix: ensure WebSocket is subscribed before operations by @xingyaoww in https://github.com/OpenHands/software-agent-sdk/pull/1791
refactor: remove redundant shutdown reconcile call by @xingyaoww in https://github.com/OpenHands/software-agent-sdk/pull/1820
Fix: Add EXAMPLE_COST marker to critic example by @xingyaoww in https://github.com/OpenHands/software-agent-sdk/pull/1830

Full Changelog: https://github.com/OpenHands/software-agent-sdk/compare/v1.9.1...v1.10.0

Open source →

1.9.121 Jan 2026

Add 05 skills/plugins examples to test-examples by @enyst in https://github.com/OpenHands/software-agent-sdk/pull/1773

What's Changed

Add 05 skills/plugins examples to test-examples by @enyst in https://github.com/OpenHands/software-agent-sdk/pull/1773
Release v1.9.0 by @all-hands-bot in https://github.com/OpenHands/software-agent-sdk/pull/1775
Added more fixes for parsing discriminated unions by @tofarr in https://github.com/OpenHands/software-agent-sdk/pull/1779

Full Changelog: https://github.com/OpenHands/software-agent-sdk/compare/v1.9.0...v1.9.1

Open source →

1.9.021 Jan 2026

Added support for experimental critic model

https://github.com/user-attachments/assets/a8e188ee-ff82-4068-9c8b-5e7e9ba42508

What's Changed

Propagate Datadog conversation logging flag to evaluation dispatch by @simonrosenberg in https://github.com/OpenHands/software-agent-sdk/pull/1703
Release v1.8.2 by @all-hands-bot in https://github.com/OpenHands/software-agent-sdk/pull/1722
Add swebenchmultimodal support to run-eval workflow by @juanmichelini in https://github.com/OpenHands/software-agent-sdk/pull/1659
feat: add multiswebench support to run-eval workflow by @juanmichelini in https://github.com/OpenHands/software-agent-sdk/pull/1693
feat: load skills from agent server by @hieptl in https://github.com/OpenHands/software-agent-sdk/pull/1729
Add GPT-5.2 high reasoning to model configurations by @neubig in https://github.com/OpenHands/software-agent-sdk/pull/1735
Replace mocked MCP tests with real integration tests by @neubig in https://github.com/OpenHands/software-agent-sdk/pull/1678
feat(plugin): Add Plugin.fetch() for remote plugin fetching and caching by @jpshackelford in https://github.com/OpenHands/software-agent-sdk/pull/1647
Fix event ordering in RemoteEventsList by inserting events sorted by timestamp by @neubig in https://github.com/OpenHands/software-agent-sdk/pull/1737
fix: move async operations outside sync lock in event subscription by @neubig in https://github.com/OpenHands/software-agent-sdk/pull/1732
Remove unsupported models by @simonrosenberg in https://github.com/OpenHands/software-agent-sdk/pull/1734
refactor(hooks): add typed fields to HookConfig for better type safety by @xingyaoww in https://github.com/OpenHands/software-agent-sdk/pull/1726
Add Marketplace datamodel for marketplace.json support by @neubig in https://github.com/OpenHands/software-agent-sdk/pull/1744
Remove push_to_index option from run-eval.yml workflow by @juanmichelini in https://github.com/OpenHands/software-agent-sdk/pull/1750
feat: add Gemini 3 models to reasoning effort supported list by @Chesars in https://github.com/OpenHands/software-agent-sdk/pull/1752
Fix DiscriminatedUnionEnvParser to use single parser when only one kind exists by @tofarr in https://github.com/OpenHands/software-agent-sdk/pull/1741
Add GitHub repository URLs to PyPI package metadata by @xingyaoww in https://github.com/OpenHands/software-agent-sdk/pull/1753
Fix cache tag truncation with ports and latest suffix by @simonrosenberg in https://github.com/OpenHands/software-agent-sdk/pull/1626
docs: rewrite CONTRIBUTING around SDK architecture by @enyst in https://github.com/OpenHands/software-agent-sdk/pull/1755
Fix CORS to allow DOCKER_HOST_ADDR for remote browser access by @shanemort1982 in https://github.com/OpenHands/software-agent-sdk/pull/1466
refactor(examples): use .from_dict() for hooks example by @xingyaoww in https://github.com/OpenHands/software-agent-sdk/pull/1746
fix: Add JSON formatter for uvicorn access logs by @neubig in https://github.com/OpenHands/software-agent-sdk/pull/1733
Fix unhandled ConversationRunError causing non-JSON logs by @neubig in https://github.com/OpenHands/software-agent-sdk/pull/1680
remove gpt-5-mini from list of models to evaluate for openhands index by @simonrosenberg in https://github.com/OpenHands/software-agent-sdk/pull/1766
Add API-Based Critic for Real-Time Agent Action Evaluation (Experimental) by @xingyaoww in https://github.com/OpenHands/software-agent-sdk/pull/1269
feat: Add .pr/ directory convention and auto-cleanup workflow by @jpshackelford in https://github.com/OpenHands/software-agent-sdk/pull/1764
feat(delegate): Add create_sub_visualizer method for custom sub-agent visualization by @malhotra5 in https://github.com/OpenHands/software-agent-sdk/pull/1767
Add minimax/MiniMax-M2.1 to resolve_model_config.py by @juanmichelini in https://github.com/OpenHands/software-agent-sdk/pull/1757
feat: Support full class names in DiscriminatedUnionEnvParser by @tofarr in https://github.com/OpenHands/software-agent-sdk/pull/1768
Update idle time on bash, git, and file operations by @neubig in https://github.com/OpenHands/software-agent-sdk/pull/1770
Add set_security_analyzer abstract method to BaseConversation by @malhotra5 in https://github.com/OpenHands/software-agent-sdk/pull/1772

New Contributors

@shanemort1982 made their first contribution in https://github.com/OpenHands/software-agent-sdk/pull/1466

Full Changelog: https://github.com/OpenHands/software-agent-sdk/compare/v1.8.2...v1.9.0

Open source →

1.8.214 Jan 2026

Release v1.8.1 by @all-hands-bot in https://github.com/OpenHands/software-agent-sdk/pull/1670

What's Changed

Release v1.8.1 by @all-hands-bot in https://github.com/OpenHands/software-agent-sdk/pull/1670
feat: auto-trigger PR review on new ready-for-review PRs + fix duplicate comments by @xingyaoww in https://github.com/OpenHands/software-agent-sdk/pull/1669
Add --no-pager instruction for git commands in system prompt by @xingyaoww in https://github.com/OpenHands/software-agent-sdk/pull/1673
fix: mask LLM_API_KEY and GITHUB_TOKEN secrets in PR review agent by @xingyaoww in https://github.com/OpenHands/software-agent-sdk/pull/1675
Add per-test timeout to prevent hanging example scripts by @xingyaoww in https://github.com/OpenHands/software-agent-sdk/pull/1666
Fix PR review agent multi-line suggestion alignment by @xingyaoww in https://github.com/OpenHands/software-agent-sdk/pull/1677
Add visualize() for ConversationErrorEvent by @enyst in https://github.com/OpenHands/software-agent-sdk/pull/1686
Fix useless logging in sockets.py by @neubig in https://github.com/OpenHands/software-agent-sdk/pull/1679
Fix conversation restore failing due to secret serialization/deserialization mismatch by @tofarr in https://github.com/OpenHands/software-agent-sdk/pull/1672
fix(critic): Accept MessageEvent as valid finish signal in AgentFinishedCritic by @neubig in https://github.com/OpenHands/software-agent-sdk/pull/1695
fix: add file-based locking to EventLog.append() by @ixchio in https://github.com/OpenHands/software-agent-sdk/pull/1614
feat(pr-review): Add priority labels to review comments by @xingyaoww in https://github.com/OpenHands/software-agent-sdk/pull/1696
feat: support custom volumes mounting for DockerWorkspace by @Co1lin in https://github.com/OpenHands/software-agent-sdk/pull/1618
Add AGENTS.md and link repo.md by @enyst in https://github.com/OpenHands/software-agent-sdk/pull/1690
Condenser integration tests by @csmith49 in https://github.com/OpenHands/software-agent-sdk/pull/1652
Fix: Disable streaming for sub-agents in delegate tool by @malhotra5 in https://github.com/OpenHands/software-agent-sdk/pull/1705
Fix deepseek-v3.2-reasoner model name to use valid API identifier by @neubig in https://github.com/OpenHands/software-agent-sdk/pull/1707
fix(condenser): Forgetting range calculation interprets keep_first correctly by @csmith49 in https://github.com/OpenHands/software-agent-sdk/pull/1708
llm: log failed Responses calls with request context by @enyst in https://github.com/OpenHands/software-agent-sdk/pull/1684
Fix verify method to include builtin tools in event check by @malhotra5 in https://github.com/OpenHands/software-agent-sdk/pull/1710
Optimize docker-like behaviour in apptainer with better code quality by @adityasoni9998 in https://github.com/OpenHands/software-agent-sdk/pull/1711
Remove .openhands/skills/repo.md symlink by @enyst in https://github.com/OpenHands/software-agent-sdk/pull/1709
Add crash diagnostics logging to agent server by @neubig in https://github.com/OpenHands/software-agent-sdk/pull/1689
Fix PR review bot to diff against current base commit by @enyst in https://github.com/OpenHands/software-agent-sdk/pull/1685
Add ARG USERNAME to base-image stage for podman by @cbagwell in https://github.com/OpenHands/software-agent-sdk/pull/1691
Revert "fix(critic): Accept MessageEvent as valid finish signal in AgentFinishedCritic" by @xingyaoww in https://github.com/OpenHands/software-agent-sdk/pull/1715
Add configurable startup wait for APIRemoteWorkspace by @simonrosenberg in https://github.com/OpenHands/software-agent-sdk/pull/1713
Add user tracking to evaluation workflow by @juanmichelini in https://github.com/OpenHands/software-agent-sdk/pull/1716
fix: route qwen-3-coder through Fireworks provider by @neubig in https://github.com/OpenHands/software-agent-sdk/pull/1720

New Contributors

@Co1lin made their first contribution in https://github.com/OpenHands/software-agent-sdk/pull/1618

Full Changelog: https://github.com/OpenHands/software-agent-sdk/compare/v1.8.1...v1.8.2

Open source →

1.8.19 Jan 2026

Fix the failed PyPI release on v1.8.0

What's Changed

Release v1.8.0 by @all-hands-bot in https://github.com/OpenHands/software-agent-sdk/pull/1663
fix: make publish workflow idempotent for re-runs by @xingyaoww in https://github.com/OpenHands/software-agent-sdk/pull/1667
feat: Add inline review comments for PR review workflow by @xingyaoww in https://github.com/OpenHands/software-agent-sdk/pull/1654
fix(tests): Enable accurate coverage reporting with pytest-xdist by @jpshackelford in https://github.com/OpenHands/software-agent-sdk/pull/1658

Full Changelog: https://github.com/OpenHands/software-agent-sdk/compare/v1.8.0...v1.8.1

Open source →

1.8.09 Jan 2026

Support for AgentSkills standard

Major Feature

Support for AgentSkills standard

What's Changed

Add push_to_index dispatch flag by @simonrosenberg in https://github.com/OpenHands/software-agent-sdk/pull/1588
feat: support .mcp.json for MCP server configuration by @neubig in https://github.com/OpenHands/software-agent-sdk/pull/1481
Release v1.7.4 by @all-hands-bot in https://github.com/OpenHands/software-agent-sdk/pull/1595
feat: support resource directories (scripts/, references/, assets/) by @neubig in https://github.com/OpenHands/software-agent-sdk/pull/1482
Fix stats being reset when resuming a conversation by @xingyaoww in https://github.com/OpenHands/software-agent-sdk/pull/1591
Add dynamic run name title to Run Eval workflow by @juanmichelini in https://github.com/OpenHands/software-agent-sdk/pull/1612
refactor: remove reconciliation methods, use runtime agent directly by @enyst in https://github.com/OpenHands/software-agent-sdk/pull/1542
fix(git): skip remote check for repos with no commits by @neubig in https://github.com/OpenHands/software-agent-sdk/pull/1606
Add code review microagent based on xingyaoww's review history by @xingyaoww in https://github.com/OpenHands/software-agent-sdk/pull/1150
feat(sdk): complete hooks implementation with additional context and stop hook by @ak684 in https://github.com/OpenHands/software-agent-sdk/pull/1547
feat: Add Plugin data model and basic loading from directories by @neubig in https://github.com/OpenHands/software-agent-sdk/pull/1611
feat: Add plugin loading example demonstrating skills, hooks, and MCP by @neubig in https://github.com/OpenHands/software-agent-sdk/pull/1616
Expose runtime_resource_factor in run-eval by @simonrosenberg in https://github.com/OpenHands/software-agent-sdk/pull/1605
feat: add skill validation improvements by @neubig in https://github.com/OpenHands/software-agent-sdk/pull/1597
Add skills support to PR review example with /codereview and /codereview-roasted by @xingyaoww in https://github.com/OpenHands/software-agent-sdk/pull/1617
feat: add to_prompt() for XML skill prompt generation by @neubig in https://github.com/OpenHands/software-agent-sdk/pull/1483
Fix poetry lock command for Poetry 2.x compatibility by @xingyaoww in https://github.com/OpenHands/software-agent-sdk/pull/1622
feat: Add skills example and include skill location in agent prompt by @neubig in https://github.com/OpenHands/software-agent-sdk/pull/1599
Multi-step integration tests by @csmith49 in https://github.com/OpenHands/software-agent-sdk/pull/1613
Add action summary feature for tool calls by @xingyaoww in https://github.com/OpenHands/software-agent-sdk/pull/1339
Add expected index models to resolve_model_config.py by @juanmichelini in https://github.com/OpenHands/software-agent-sdk/pull/1497
Revert "Expose runtime_resource_factor in run-eval" by @simonrosenberg in https://github.com/OpenHands/software-agent-sdk/pull/1627
fix: resolve UnsupportedFieldAttributeWarning for secret_registry field by @neubig in https://github.com/OpenHands/software-agent-sdk/pull/1625
Fix context window displaying as 0 when resuming conversation by @xingyaoww in https://github.com/OpenHands/software-agent-sdk/pull/1590
fix: Update PR review workflow to use correct repository name by @xingyaoww in https://github.com/OpenHands/software-agent-sdk/pull/1623
Fix AGENTS.md not being loaded when skills directories don't exist by @xingyaoww in https://github.com/OpenHands/software-agent-sdk/pull/1624
fix: Pass agent_context at Agent initialization in PR review script by @xingyaoww in https://github.com/OpenHands/software-agent-sdk/pull/1632
Allow over-riding system-level configurations from apptainer.conf in apptainer workspace by @adityasoni9998 in https://github.com/OpenHands/software-agent-sdk/pull/1628
feat(llm): Set default LLM timeout to 5 minutes (300s) by @neubig in https://github.com/OpenHands/software-agent-sdk/pull/1638
feat(pr_review): send complete git diff to agent in first message by @xingyaoww in https://github.com/OpenHands/software-agent-sdk/pull/1639
fix: Auto-add skills to_prompt() output to system prompt by @neubig in https://github.com/OpenHands/software-agent-sdk/pull/1642
feat: Add include_default_tools option to control built-in tools by @xingyaoww in https://github.com/OpenHands/software-agent-sdk/pull/1594
fix: remove duplicate example numbers in 02_remote_agent_server by @xingyaoww in https://github.com/OpenHands/software-agent-sdk/pull/1637
Use AGENTS.md instead of repo.md in SDK examples/system prompt by @enyst in https://github.com/OpenHands/software-agent-sdk/pull/1648
feat: Add public sandbox_id field to OpenHandsCloudWorkspace for resuming existing sandboxes by @malhotra5 in https://github.com/OpenHands/software-agent-sdk/pull/1603
docs: Update environment configuration for secret encryption in Agent Server by @jagzmz in https://github.com/OpenHands/software-agent-sdk/pull/1465
feat: add utility to convert legacy OpenHands skills to AgentSkills format by @xingyaoww in https://github.com/OpenHands/software-agent-sdk/pull/1643
feat: Implement AgentSkills progressive disclosure for SKILL.md files by @xingyaoww in https://github.com/OpenHands/software-agent-sdk/pull/1644
Add alive property to RemoteWorkspace by @malhotra5 in https://github.com/OpenHands/software-agent-sdk/pull/1655
Expose max_retries input and dispatch to eval by @simonrosenberg in https://github.com/OpenHands/software-agent-sdk/pull/1641

New Contributors

@jagzmz made their first contribution in https://github.com/OpenHands/software-agent-sdk/pull/1465

Full Changelog: https://github.com/OpenHands/software-agent-sdk/compare/v1.7.4...v1.8.0

Open source →

1.7.45 Jan 2026Deprecated

build(deps): Bump libtmux to >=0.53.0 for v0.51.0 hard deprecations by @tony in https://github.com/OpenHands/software-agent-sdk/pull/1561

What's Changed

Release v1.7.3 by @all-hands-bot in https://github.com/OpenHands/software-agent-sdk/pull/1551
Update OpenHands version bump workflow to include all required steps by @malhotra5 in https://github.com/OpenHands/software-agent-sdk/pull/1553
fix: Wait for PyPI propagation and add Slack notification to release workflow by @xingyaoww in https://github.com/OpenHands/software-agent-sdk/pull/1552
feat: support SKILL.md file convention and name validation by @neubig in https://github.com/OpenHands/software-agent-sdk/pull/1480
feat: add pause() and resume() methods to workspace classes by @ixchio in https://github.com/OpenHands/software-agent-sdk/pull/1539
build(deps): Bump libtmux to >=0.53.0 for v0.51.0 hard deprecations by @tony in https://github.com/OpenHands/software-agent-sdk/pull/1561
feat: add DeepSeek v3.2 model support by @ixchio in https://github.com/OpenHands/software-agent-sdk/pull/1564
refactor: consolidate duplicated blocking wait logic in RemoteConversation.run() + remove defensive if else by @simonrosenberg in https://github.com/OpenHands/software-agent-sdk/pull/1569
Fix DiscriminatedUnionMixin issues with Pydantic C bindings by @tofarr in https://github.com/OpenHands/software-agent-sdk/pull/1555
Add test to verify working_dir standardization (closes #211) by @xingyaoww in https://github.com/OpenHands/software-agent-sdk/pull/1576
Re-add discriminator field to discriminated union JSON schema by @tofarr in https://github.com/OpenHands/software-agent-sdk/pull/1578
fix: Cap thinking budget below max_tokens by @csmith49 in https://github.com/OpenHands/software-agent-sdk/pull/1580
Relax needs-triage duration by @enyst in https://github.com/OpenHands/software-agent-sdk/pull/1567
Fix RemoteConversation polling on terminal errors by @simonrosenberg in https://github.com/OpenHands/software-agent-sdk/pull/1572
More test fixes by @tofarr in https://github.com/OpenHands/software-agent-sdk/pull/1581
remove custom_llm_provider from LLM config by @enyst in https://github.com/OpenHands/software-agent-sdk/pull/1583
Update cryptography to 46.0.3 by @enyst in https://github.com/OpenHands/software-agent-sdk/pull/1585
feat: relax tool matching on resume by @ixchio in https://github.com/OpenHands/software-agent-sdk/pull/1538
Set initial execution status to error if it was running by @tofarr in https://github.com/OpenHands/software-agent-sdk/pull/1554
More efficient websocket based implementation for async bash execution by @tofarr in https://github.com/OpenHands/software-agent-sdk/pull/1587
feat: Create conversation when provided ID doesn't exist by @malhotra5 in https://github.com/OpenHands/software-agent-sdk/pull/1579
fix(condenser): Retry on empty condensation by @csmith49 in https://github.com/OpenHands/software-agent-sdk/pull/1577
Bump pydantic by @tofarr in https://github.com/OpenHands/software-agent-sdk/pull/1593

New Contributors

@tony made their first contribution in https://github.com/OpenHands/software-agent-sdk/pull/1561

Full Changelog: https://github.com/OpenHands/software-agent-sdk/compare/v1.7.3...v1.7.4

Open source →

1.7.330 Dec 2025

feat: Add support for custom tools with remote agent server by @xingyaoww in https://github.com/OpenHands/software-agent-sdk/pull/1383

What's Changed

feat: Add support for custom tools with remote agent server by @xingyaoww in https://github.com/OpenHands/software-agent-sdk/pull/1383
Release v1.7.2 by @all-hands-bot in https://github.com/OpenHands/software-agent-sdk/pull/1530
Add ApptainerWorkspace implementation for rootless container support by @neubig in https://github.com/OpenHands/software-agent-sdk/pull/892
Add GPT-5 preset using ApplyPatchTool (opt-in) by @enyst in https://github.com/OpenHands/software-agent-sdk/pull/1462
feat: add openrouter/minimax-m2 to SEND_REASONING_CONTENT_PATTERNS by @neubig in https://github.com/OpenHands/software-agent-sdk/pull/1152
Add forward_env support to APIRemoteWorkspace by @simonrosenberg in https://github.com/OpenHands/software-agent-sdk/pull/1540
Raise ValueError when condensing 0 events by @xingyaoww in https://github.com/OpenHands/software-agent-sdk/pull/1541
Propagate remote conversation errors by @simonrosenberg in https://github.com/OpenHands/software-agent-sdk/pull/1546
Fix save_meta() not being called on conversation creation by @tofarr in https://github.com/OpenHands/software-agent-sdk/pull/1548
Fix ValueError from tool execution causing conversation to fail by @xingyaoww in https://github.com/OpenHands/software-agent-sdk/pull/1550

Full Changelog: https://github.com/OpenHands/software-agent-sdk/compare/v1.7.2...v1.7.3

Open source →

1.7.229 Dec 2025

Release v1.7.1 by @all-hands-bot in https://github.com/OpenHands/software-agent-sdk/pull/1513

What's Changed

Release v1.7.1 by @all-hands-bot in https://github.com/OpenHands/software-agent-sdk/pull/1513
Load model-specific repo instructions (CLAUDE.md for Claude only, GEMINI.md for Gemini only) by @enyst in https://github.com/OpenHands/software-agent-sdk/pull/1328
Fix max iterations reached: Set ERROR status with clear message by @simonrosenberg in https://github.com/OpenHands/software-agent-sdk/pull/1511
Save screenshots from browser tool by @hamiltop in https://github.com/OpenHands/software-agent-sdk/pull/1443
Use ALLHANDS_BOT_GITHUB_PAT for bot-like workflows by @xingyaoww in https://github.com/OpenHands/software-agent-sdk/pull/1519
Fix TmuxTerminal.close() to handle dead session exceptions by @xingyaoww in https://github.com/OpenHands/software-agent-sdk/pull/1517
ci(integration): update integration LLM matrix to gpt-5.2-codex by @enyst in https://github.com/OpenHands/software-agent-sdk/pull/1503
Add automated version bump PRs for downstream repos by @xingyaoww in https://github.com/OpenHands/software-agent-sdk/pull/1520
feat: add early stopping/pruner for behavior tests cost optimization by @ixchio in https://github.com/OpenHands/software-agent-sdk/pull/1433
Fix condensation for 0 events by adjusting default parameters by @xingyaoww in https://github.com/OpenHands/software-agent-sdk/pull/1523
Add pull request template by @neubig in https://github.com/OpenHands/software-agent-sdk/pull/1528
Revert "ci(integration): update integration LLM matrix to gpt-5.2-codex" by @xingyaoww in https://github.com/OpenHands/software-agent-sdk/pull/1524

New Contributors

@all-hands-bot made their first contribution in https://github.com/OpenHands/software-agent-sdk/pull/1513
@hamiltop made their first contribution in https://github.com/OpenHands/software-agent-sdk/pull/1443
@ixchio made their first contribution in https://github.com/OpenHands/software-agent-sdk/pull/1433

Full Changelog: https://github.com/OpenHands/software-agent-sdk/compare/v1.7.1...v1.7.2

Open source →

1.7.125 Dec 2025

feat: add AgentSkills standard fields to Skill model by @neubig in https://github.com/OpenHands/software-agent-sdk/pull/1479

What's Changed

feat: add AgentSkills standard fields to Skill model by @neubig in https://github.com/OpenHands/software-agent-sdk/pull/1479
Release v1.7.0 by @xingyaoww in https://github.com/OpenHands/software-agent-sdk/pull/1486
Fix SDK_SHA reference in run-eval.yml Comment on PR step by @simonrosenberg in https://github.com/OpenHands/software-agent-sdk/pull/1498
Add optional parallelization parameters to run-eval workflow by @simonrosenberg in https://github.com/OpenHands/software-agent-sdk/pull/1499
Add compatibility fixes for Nemotron and similar models by @pankajarm in https://github.com/OpenHands/software-agent-sdk/pull/1470
feat(condenser): Token-aware condensation in LLMSummarizingCondenser by @csmith49 in https://github.com/OpenHands/software-agent-sdk/pull/1380
fix(condenser): Tool-call aware condensation by @csmith49 in https://github.com/OpenHands/software-agent-sdk/pull/1412
feat(localFileStore): Add Cache Layer to LocalFileStore to Avoid I/O Overhead from Read Operations by @CLFutureX in https://github.com/OpenHands/software-agent-sdk/pull/1274
feat: Add hooks system for event interception by @ak684 in https://github.com/OpenHands/software-agent-sdk/pull/1467
fix: handle union types in py_type for mcp schema conversion by @hieptl in https://github.com/OpenHands/software-agent-sdk/pull/1509
Remove tool preambles from gpt-5-codex by @enyst in https://github.com/OpenHands/software-agent-sdk/pull/1512
fix(condenser): Tool-loop aware condensation by @csmith49 in https://github.com/OpenHands/software-agent-sdk/pull/1508

New Contributors

@pankajarm made their first contribution in https://github.com/OpenHands/software-agent-sdk/pull/1470

Full Changelog: https://github.com/OpenHands/software-agent-sdk/compare/v1.7.0...v1.7.1

Open source →

1.7.023 Dec 2025

Release 1.6.0 by @xingyaoww in https://github.com/OpenHands/software-agent-sdk/pull/1404

What's Changed

Release 1.6.0 by @xingyaoww in https://github.com/OpenHands/software-agent-sdk/pull/1404
Make thinking signatures optional by @cbagwell in https://github.com/OpenHands/software-agent-sdk/pull/1394
Make integration-test label run integration tests only by @ryanhoangt in https://github.com/OpenHands/software-agent-sdk/pull/1416
make eval_limit param optional by @simonrosenberg in https://github.com/OpenHands/software-agent-sdk/pull/1419
Exclude '*mini' models from prompt_cache_retention by @enyst in https://github.com/OpenHands/software-agent-sdk/pull/1345
Add load_project_skills helper function by @xingyaoww in https://github.com/OpenHands/software-agent-sdk/pull/1420
feat(AsyncExecutor): AsyncExecutor allows adding tasks after shutdown, which may cause unpredictable exceptions. by @CLFutureX in https://github.com/OpenHands/software-agent-sdk/pull/1092
fix end span in local conversation run by @Rainhunter13 in https://github.com/OpenHands/software-agent-sdk/pull/1432
Add iterative refinement example for COBOL to Java refactoring by @neubig in https://github.com/OpenHands/software-agent-sdk/pull/1415
Make cronjob run integration test only by @ryanhoangt in https://github.com/OpenHands/software-agent-sdk/pull/1430
Add commit0 to CI evaluation workflow by @simonrosenberg in https://github.com/OpenHands/software-agent-sdk/pull/1429
delete image when docker workspace closes by @madhur-tandon in https://github.com/OpenHands/software-agent-sdk/pull/1424
Include persistence_dir in ConversationRunError for better debugging by @xingyaoww in https://github.com/OpenHands/software-agent-sdk/pull/1437
Fix JSON encoding to preserve non-ASCII characters in model_dump_json by @LuneZ99 in https://github.com/OpenHands/software-agent-sdk/pull/1431
refactor: add support for description field in sdk secrets by @hieptl in https://github.com/OpenHands/software-agent-sdk/pull/1428
include j2 templates in tools package by @madhur-tandon in https://github.com/OpenHands/software-agent-sdk/pull/1426
Fix: throw error when viewing PDF files instead of crashing by @xingyaoww in https://github.com/OpenHands/software-agent-sdk/pull/1423
Update litellm to 1.80.10 and add tests for Gemini 2.5 thinking blocks by @xingyaoww in https://github.com/OpenHands/software-agent-sdk/pull/1413
Suppress LiteLLM INFO logs to clean up output by @xingyaoww in https://github.com/OpenHands/software-agent-sdk/pull/1369
Make stuck detection thresholds configurable by @neubig in https://github.com/OpenHands/software-agent-sdk/pull/763
feat: Add OpenHandsCloudWorkspace for OpenHands Cloud API by @xingyaoww in https://github.com/OpenHands/software-agent-sdk/pull/1442
Add PostHog Debugging Example by @neubig in https://github.com/OpenHands/software-agent-sdk/pull/1202
Public APIs return OpenHands types not LiteLLM types - Part 2: SystemPromptEvent by @jpshackelford in https://github.com/OpenHands/software-agent-sdk/pull/495
refactor: Add LookupSecret tests for AgentContext.get_secret_infos() by @hieptl in https://github.com/OpenHands/software-agent-sdk/pull/1456
Add nova-2-lite to model_features by @BalasubramanyamEvani in https://github.com/OpenHands/software-agent-sdk/pull/1455
fix: disable scheduled workflows from running in forks by @Chesars in https://github.com/OpenHands/software-agent-sdk/pull/1447
Add skill for writing behavior tests by @ryanhoangt in https://github.com/OpenHands/software-agent-sdk/pull/1459
feat: Add configurable security policy support to Agent by @neubig in https://github.com/OpenHands/software-agent-sdk/pull/427
Fix batch atomicity when condensation forgets ObservationEvents by @xingyaoww in https://github.com/OpenHands/software-agent-sdk/pull/1450
fix: add openrouter/minimax to FORCE_STRING_SERIALIZER_MODELS by @neubig in https://github.com/OpenHands/software-agent-sdk/pull/1448
Fix verified_models by @enyst in https://github.com/OpenHands/software-agent-sdk/pull/1458
fix: deepseek reasoner model not part of send_reasoning_content_support models (Closes #1343) by @BalasubramanyamEvani in https://github.com/OpenHands/software-agent-sdk/pull/1464
Implement gemini-style file editing tools by @neubig in https://github.com/OpenHands/software-agent-sdk/pull/1199
terminal: populate TerminalObservation.exit_code (fixes #1468) by @enyst in https://github.com/OpenHands/software-agent-sdk/pull/1469
Fix RemoteConversation.run() timeout issues with async polling by @xingyaoww in https://github.com/OpenHands/software-agent-sdk/pull/1444
Add accumulated token usage to integration tests output and GitHub CI comments by @xingyaoww in https://github.com/OpenHands/software-agent-sdk/pull/1471
Fix prepare-release workflow to use ALLHANDS_BOT_GITHUB_PAT by @xingyaoww in https://github.com/OpenHands/software-agent-sdk/pull/1485
Fix prepare-release workflow to use correct label names by @xingyaoww in https://github.com/OpenHands/software-agent-sdk/pull/1487

New Contributors

@cbagwell made their first contribution in https://github.com/OpenHands/software-agent-sdk/pull/1394
@Rainhunter13 made their first contribution in https://github.com/OpenHands/software-agent-sdk/pull/1432
@madhur-tandon made their first contribution in https://github.com/OpenHands/software-agent-sdk/pull/1424
@BalasubramanyamEvani made their first contribution in https://github.com/OpenHands/software-agent-sdk/pull/1455

Full Changelog: https://github.com/OpenHands/software-agent-sdk/compare/v1.6.0...v1.7.0

Open source →

1.6.016 Dec 2025

Release 1.5.2 by @xingyaoww in https://github.com/OpenHands/software-agent-sdk/pull/1382

What's Changed

Release 1.5.2 by @xingyaoww in https://github.com/OpenHands/software-agent-sdk/pull/1382
feat: allow xhigh reasoning_effort by @enyst in https://github.com/OpenHands/software-agent-sdk/pull/1388
Add page_iterator utility and refactor resend_all functionality by @tofarr in https://github.com/OpenHands/software-agent-sdk/pull/1386
ALL-4485 Added deleting state for conversations for use in webhook. by @tofarr in https://github.com/OpenHands/software-agent-sdk/pull/1391
fix: use simple delay for shell init instead of PTY-based detection by @ak684 in https://github.com/OpenHands/software-agent-sdk/pull/1387
Add instance id input to run-eval dispatch by @simonrosenberg in https://github.com/OpenHands/software-agent-sdk/pull/1389
Update lmnr from 0.7.20 to 0.7.24 by @xingyaoww in https://github.com/OpenHands/software-agent-sdk/pull/1398
Add self-documentation lookup capability to system prompt by @xingyaoww in https://github.com/OpenHands/software-agent-sdk/pull/1396
ALL-4516 Fix selfish coroutines in event service state lock operations by @tofarr in https://github.com/OpenHands/software-agent-sdk/pull/1400
docs(readme): update Slack link to openhands.dev domain by @enyst in https://github.com/OpenHands/software-agent-sdk/pull/1397
ALL-4518 Fix agent deep copy to preserve secrets using expose_secrets by @tofarr in https://github.com/OpenHands/software-agent-sdk/pull/1403
Add MEMORY section to system prompt for skills by @xingyaoww in https://github.com/OpenHands/software-agent-sdk/pull/600
Support model-family and model-variant system prompts by @ryanhoangt in https://github.com/OpenHands/software-agent-sdk/pull/1348
Add automated release workflows by @xingyaoww in https://github.com/OpenHands/software-agent-sdk/pull/1405

Full Changelog: https://github.com/OpenHands/software-agent-sdk/compare/v1.5.2...v1.6.0

Open source →

1.5.211 Dec 2025

Release 1.5.1 by @xingyaoww in https://github.com/OpenHands/software-agent-sdk/pull/1379

What's Changed

Release 1.5.1 by @xingyaoww in https://github.com/OpenHands/software-agent-sdk/pull/1379
ALL-4426 Optimize agent tool initialization with parallel execution by @tofarr in https://github.com/OpenHands/software-agent-sdk/pull/1374
ALL-4426 Parallelize service startup and shutdown in api_lifespan by @tofarr in https://github.com/OpenHands/software-agent-sdk/pull/1375
ALL-4426 Fix: Run conversation webhook notifications in background with error logging by @tofarr in https://github.com/OpenHands/software-agent-sdk/pull/1373
fix: update respond_to_confirmation API to Invoke reject_pending_actions on rejection by @hieptl in https://github.com/OpenHands/software-agent-sdk/pull/1377

Full Changelog: https://github.com/OpenHands/software-agent-sdk/compare/v1.5.1...v1.5.2

Open source →

1.5.110 Dec 2025

Release 1.5.0 by @xingyaoww in https://github.com/OpenHands/software-agent-sdk/pull/1370

What's Changed

Release 1.5.0 by @xingyaoww in https://github.com/OpenHands/software-agent-sdk/pull/1370
Fix agent reconciliation to allow agent_context updates by @jpshackelford in https://github.com/OpenHands/software-agent-sdk/pull/1371
Add GPT 5.1 codex models to verified models list by @Chesars in https://github.com/OpenHands/software-agent-sdk/pull/1358
Add agent behavior tests to integration tests by @xingyaoww in https://github.com/OpenHands/software-agent-sdk/pull/1321
Add condense method to conversation class with ForceCondenser by @malhotra5 in https://github.com/OpenHands/software-agent-sdk/pull/1368

Full Changelog: https://github.com/OpenHands/software-agent-sdk/compare/v1.5.0...v1.5.1

Open source →

1.5.09 Dec 2025

Release 1.4.1 by @xingyaoww in https://github.com/OpenHands/software-agent-sdk/pull/1322

Support Latest Devstral
Add GPT-5.1 Models
Couple bug fixes

What's Changed

Release 1.4.1 by @xingyaoww in https://github.com/OpenHands/software-agent-sdk/pull/1322
Fix: Enable vision for messages containing ImageContent by @ramarlina in https://github.com/OpenHands/software-agent-sdk/pull/1279
APP-190 Fix PyInstaller GCC 14 compatibility for Chromium by @tofarr in https://github.com/OpenHands/software-agent-sdk/pull/1327
Add Windows browser path checks to Chromium detection by @SmartManoj in https://github.com/OpenHands/software-agent-sdk/pull/1014
Fix server-side automatic string-to-StaticSecret conversion (v1.2.0 compatible) by @rbren in https://github.com/OpenHands/software-agent-sdk/pull/1234
Add issue triage workflow automation by @neubig in https://github.com/OpenHands/software-agent-sdk/pull/1300
fix: create parent directories in observation truncation by @simonrosenberg in https://github.com/OpenHands/software-agent-sdk/pull/1347
Fix base_url handling for openhands/ provider with explicit None value by @xingyaoww in https://github.com/OpenHands/software-agent-sdk/pull/1341
fix: respect enable_encrypted_reasoning flag in Responses API by @mr-karan in https://github.com/OpenHands/software-agent-sdk/pull/1208
Add helpful warning when context window exceeded without condenser by @xingyaoww in https://github.com/OpenHands/software-agent-sdk/pull/1354
feat(llm):Enhance retry listeners and metric collection when retrying LLM calls due to exceptions. by @CLFutureX in https://github.com/OpenHands/software-agent-sdk/pull/1214
Add benchmark selection parameter to evaluation workflow by @simonrosenberg in https://github.com/OpenHands/software-agent-sdk/pull/1294
fix: enable agent awareness of available secrets by @hieptl in https://github.com/OpenHands/software-agent-sdk/pull/1335
Update SWE-bench score to 77.6% by @neubig in https://github.com/OpenHands/software-agent-sdk/pull/1362
Add devstral-2512 and devstral-medium-2512 as verified models by @xingyaoww in https://github.com/OpenHands/software-agent-sdk/pull/1364
Add no-mocking instructions to system prompt by @ak684 in https://github.com/OpenHands/software-agent-sdk/pull/1357
Fix MCP tool observation crash when tool returns a list by @malhotra5 in https://github.com/OpenHands/software-agent-sdk/pull/1366
Fix terminal tool crash on large env vars by @aoli-al in https://github.com/OpenHands/software-agent-sdk/pull/1346
Add GPT-5.1 models to verified model list by @xingyaoww in https://github.com/OpenHands/software-agent-sdk/pull/1365
ci: mark check-examples workflow as optional in title by @enyst in https://github.com/OpenHands/software-agent-sdk/pull/1355

New Contributors

@ramarlina made their first contribution in https://github.com/OpenHands/software-agent-sdk/pull/1279
@ak684 made their first contribution in https://github.com/OpenHands/software-agent-sdk/pull/1357
@aoli-al made their first contribution in https://github.com/OpenHands/software-agent-sdk/pull/1346

Full Changelog: https://github.com/OpenHands/software-agent-sdk/compare/v1.4.1...v1.5.0

Open source →

1.4.14 Dec 2025

Release 1.4.0 by @xingyaoww in https://github.com/OpenHands/software-agent-sdk/pull/1299

What's Changed

Release 1.4.0 by @xingyaoww in https://github.com/OpenHands/software-agent-sdk/pull/1299
Fix ConversationStateUpdateEvent to use MetricsSnapshot instead of full Metrics by @xingyaoww in https://github.com/OpenHands/software-agent-sdk/pull/1308
Fix evaluation workflow failure when release branches are deleted by @simonrosenberg in https://github.com/OpenHands/software-agent-sdk/pull/1310
Fixes where the example tools were invalid by @tofarr in https://github.com/OpenHands/software-agent-sdk/pull/1311
APP-215 Fix browser_use logging configuration interference by @tofarr in https://github.com/OpenHands/software-agent-sdk/pull/1316

Full Changelog: https://github.com/OpenHands/software-agent-sdk/compare/v1.4.0...v1.4.1

Open source →

1.4.03 Dec 2025Deprecated

[chore] Clean up deprecated service_id by @enyst in https://github.com/OpenHands/software-agent-sdk/pull/1276

What's Changed

Release 1.3.0 by @xingyaoww in https://github.com/OpenHands/software-agent-sdk/pull/1262
feat(fifolock): acquire and release Logic optimization by @CLFutureX in https://github.com/OpenHands/software-agent-sdk/pull/1251
Fix api_timeout not being used in APIRemoteWorkspace by @xingyaoww in https://github.com/OpenHands/software-agent-sdk/pull/1206
Implement streaming for Chat Completions by @xingyaoww in https://github.com/OpenHands/software-agent-sdk/pull/1270
Add load_public_skills() to load skills from public registry by @xingyaoww in https://github.com/OpenHands/software-agent-sdk/pull/1248
[chore] Clean up deprecated service_id by @enyst in https://github.com/OpenHands/software-agent-sdk/pull/1276
GPT-5.1: Add prompt cache retention; reasoning_effort tweak by @enyst in https://github.com/OpenHands/software-agent-sdk/pull/1162
Add support for loading CLAUDE.md and GEMINI.md as repo skills by @xingyaoww in https://github.com/OpenHands/software-agent-sdk/pull/1280
terminal/tmux: disable bash history expansion to avoid ! mangling by @enyst in https://github.com/OpenHands/software-agent-sdk/pull/1277
refactor: update request timeout handling when fetching secrets by @hieptl in https://github.com/OpenHands/software-agent-sdk/pull/1272
feat(tools): Add GPT-5.1 ApplyPatch tool by @enyst in https://github.com/OpenHands/software-agent-sdk/pull/1166
feat: add Streamlit viewer for completion logs by @Wangmerlyn in https://github.com/OpenHands/software-agent-sdk/pull/1284
Fix GPT-5 codex empty patches by @juanmichelini in https://github.com/OpenHands/software-agent-sdk/pull/1207
refactor: enable configuring the llm proxy base url for the openhands provider by @hieptl in https://github.com/OpenHands/software-agent-sdk/pull/1273
Enhance truncation with persistence capability by @ryanhoangt in https://github.com/OpenHands/software-agent-sdk/pull/419
Feat/tom agent integration by @XuhuiZhou in https://github.com/OpenHands/software-agent-sdk/pull/784
Add GitHub CLI installation to Dockerfile by @xingyaoww in https://github.com/OpenHands/software-agent-sdk/pull/1286
make get_closest_git_repo robust to macOS symlink paths by @enyst in https://github.com/OpenHands/software-agent-sdk/pull/1289
Simplify evaluation workflow by removing benchmarks build polling by @simonrosenberg in https://github.com/OpenHands/software-agent-sdk/pull/1267
feat: handle proxied model names with canonical model_real_name by @Wangmerlyn in https://github.com/OpenHands/software-agent-sdk/pull/1292
Add simple version tags for agent-server Docker images by @xingyaoww in https://github.com/OpenHands/software-agent-sdk/pull/1197
Fix t08_image_file_viewing test to handle FinishAction responses by @xingyaoww in https://github.com/OpenHands/software-agent-sdk/pull/1302
Enhance DiscriminatedUnionMixin error messages by @tofarr in https://github.com/OpenHands/software-agent-sdk/pull/1301
Fix system prompt to prevent automatic markdown file creation by @xingyaoww in https://github.com/OpenHands/software-agent-sdk/pull/1288
Separate DockerWorkspace and DockerDevWorkspace to fix import issue by @xingyaoww in https://github.com/OpenHands/software-agent-sdk/pull/1198

New Contributors

@Wangmerlyn made their first contribution in https://github.com/OpenHands/software-agent-sdk/pull/1284

Full Changelog: https://github.com/OpenHands/software-agent-sdk/compare/v1.3.0...v1.4.0

Open source →

1.3.025 Nov 2025Deprecated

Fix datetime.utcnow() deprecation warning by @malhotra5 in https://github.com/OpenHands/software-agent-sdk/pull/1203

Highlights

Support for Opus 4.5
Improved reliability across the execution pipeline More predictable webhook timing, stronger conversation start behavior, and refined FastAPI union handling.
Better LLM interaction and observability WebSocket-based LLM log streaming, reasoning-first messages, cleaner serialization, and a simple ask_agent API.
Stronger evaluation workflow CI eval job, improved stability, and aligned labeling tiers.
Greater extensibility and structure Predefined subagent support, event source filtering, new task status types, and import-dependency rule enforcement.

What's Changed

Release 1.2.0 by @xingyaoww in https://github.com/OpenHands/software-agent-sdk/pull/1191
Fix webhook timer reset behavior for more predictable delivery by @tofarr in https://github.com/OpenHands/software-agent-sdk/pull/1192
fix: sanitize @OpenHands only in GitHub comments (backend) by @YakshithK in https://github.com/OpenHands/software-agent-sdk/pull/1020
fix: use user-specific cache directory for Jinja2 templates by @LuneZ99 in https://github.com/OpenHands/software-agent-sdk/pull/1200
Rename ExecuteBashAction/ExecuteBashObservation to TerminalAction/TerminalObservation by @xingyaoww in https://github.com/OpenHands/software-agent-sdk/pull/1193
Fix datetime.utcnow() deprecation warning by @malhotra5 in https://github.com/OpenHands/software-agent-sdk/pull/1203
Add force_string_serializer as LLM class argument by @xingyaoww in https://github.com/OpenHands/software-agent-sdk/pull/1204
APP-156 Set skill default trigger to None by @tofarr in https://github.com/OpenHands/software-agent-sdk/pull/1212
Rename BrowserNavigateAction and BrowserTypeAction actions by @ryanhoangt in https://github.com/OpenHands/software-agent-sdk/pull/1213
Fix: more resilient conversation start by @tofarr in https://github.com/OpenHands/software-agent-sdk/pull/1216
Revert "Rename BrowserNavigateAction and BrowserTypeAction actions" by @xingyaoww in https://github.com/OpenHands/software-agent-sdk/pull/1225
Prefix MCP tool action models to avoid kind collisions by @ryanhoangt in https://github.com/OpenHands/software-agent-sdk/pull/1217
Add source parameter to /search endpoint for event filtering by @malhotra5 in https://github.com/OpenHands/software-agent-sdk/pull/1224
feat: add eval job that runs on CI by @simonrosenberg in https://github.com/OpenHands/software-agent-sdk/pull/1167
Send reasoning first in the assistant message by @enyst in https://github.com/OpenHands/software-agent-sdk/pull/1238
Remove weekly typing improvements workflow by @simonrosenberg in https://github.com/OpenHands/software-agent-sdk/pull/1243
fix: incorrect file download API endpoint in remote_workspace_mixin by @hieptl in https://github.com/OpenHands/software-agent-sdk/pull/1245
docs: Add comments about system prompt customization options by @xingyaoww in https://github.com/OpenHands/software-agent-sdk/pull/1220
Fix run examples workflow failed on schedule run & use parallel execution with pytest by @ryanhoangt in https://github.com/OpenHands/software-agent-sdk/pull/1229
feat: fix eval workflow by @simonrosenberg in https://github.com/OpenHands/software-agent-sdk/pull/1241
Fix FastAPI discriminated union patch to be called consistently by @tofarr in https://github.com/OpenHands/software-agent-sdk/pull/1256
Revert "Fix FastAPI discriminated union patch to be called consistently" by @xingyaoww in https://github.com/OpenHands/software-agent-sdk/pull/1257
Fastapi discriminated union patch2 by @tofarr in https://github.com/OpenHands/software-agent-sdk/pull/1258
feat: allow delegator to spawn pre-defined subagents by @simonrosenberg in https://github.com/OpenHands/software-agent-sdk/pull/1223
Stream LLM logs and stats updates via WebSocket for RemoteConversation by @xingyaoww in https://github.com/OpenHands/software-agent-sdk/pull/1159
Add import dependency rules enforcement by @xingyaoww in https://github.com/OpenHands/software-agent-sdk/pull/1259
Investigate Claude model names: remove unsupported dotted variants (fixes #1260) by @enyst in https://github.com/OpenHands/software-agent-sdk/pull/1261
Add TaskTrackerStatusType type alias for task tracker status by @xingyaoww in https://github.com/OpenHands/software-agent-sdk/pull/1263
Align eval labels with benchmarks build tiers (1, 50, 200) by @simonrosenberg in https://github.com/OpenHands/software-agent-sdk/pull/1254
Claude Opus 4.5 as a reasoning model with custom effort param by @juanmichelini in https://github.com/OpenHands/software-agent-sdk/pull/1250
Add ask_agent method to conversation classes for simple LLM completions by @malhotra5 in https://github.com/OpenHands/software-agent-sdk/pull/1227
Fix: Condenser LLM should not pass extra_body (Fixes #1252) by @enyst in https://github.com/OpenHands/software-agent-sdk/pull/1266

Full Changelog: https://github.com/OpenHands/software-agent-sdk/compare/v1.2.0...v1.3.0

Open source →

1.2.018 Nov 2025Deprecated

New capabilities: Critic system integration and systematic deprecation management.

Highlights

Model support for Gemini-3-Pro.
Performance improvements across the stack, including parallelized batch operations and improved LLM initialization latency.
Developer experience improvements: unified-key examples, improved README, clearer MCP timeout errors, and fixes for serialization, metadata filtering, workspace path handling, and conversation deletion.
Visualization refinements: suppressed noisy ConversationStateUpdateEvent, corrected delegation labels, and switched the default visualizer to a cleaner horizontal-rule style.
Expanded runtime/platform support: GPU-enabled Docker execution and configurable shell path for the Terminal tool.
Better interoperability and correctness: fixed LLM forwarding, UTF-8 safe telemetry logs, correct cipher restoration, and improved DEBUG-mode HTTP exception handling.
New capabilities: Critic system integration and systematic deprecation management.
New contributors joined across GPU support, telemetry improvements, and shell-path configurability.

What's Changed

fix(examples): allow all examples to run with same LLM_API_KEY by @jpshackelford in https://github.com/OpenHands/software-agent-sdk/pull/1065
v1.1.0 release by @xingyaoww in https://github.com/OpenHands/software-agent-sdk/pull/1138
Fix executable for APIRemoteWorkspace by @xingyaoww in https://github.com/OpenHands/software-agent-sdk/pull/1141
Fix metadata filtering for non-litellm_proxy providers by @neubig in https://github.com/OpenHands/software-agent-sdk/pull/1134
Skip visualizing ConversationStateUpdateEvent in default visualizer by @xingyaoww in https://github.com/OpenHands/software-agent-sdk/pull/1151
feat: Parallelize batch operations for improved performance by @tofarr in https://github.com/OpenHands/software-agent-sdk/pull/1147
Revise README for OpenHands Software Agent SDK by @rbren in https://github.com/OpenHands/software-agent-sdk/pull/1156
Fix delegation visualization labels to show correct sender/receiver by @simonrosenberg in https://github.com/OpenHands/software-agent-sdk/pull/1119
Fix litellm_extra_body LLM forwarding by @li-boxuan in https://github.com/OpenHands/software-agent-sdk/pull/1153
Fix performance bottleneck in LLM initialization (~1.2s delay) by @tofarr in https://github.com/OpenHands/software-agent-sdk/pull/1148
Add informative error message for MCP timeout by @xingyaoww in https://github.com/OpenHands/software-agent-sdk/pull/1169
Fix Path type support for workspace and persistence_dir parameters by @xingyaoww in https://github.com/OpenHands/software-agent-sdk/pull/1116
Enabling GPU support on Docker by @reecursion in https://github.com/OpenHands/software-agent-sdk/pull/1175
Improved default visualizer to use horizontal rule instead of Panel by @xingyaoww in https://github.com/OpenHands/software-agent-sdk/pull/1170
fix: add UTF-8 encoding and ensure_ascii=False for JSON telemetry logs by @LuneZ99 in https://github.com/OpenHands/software-agent-sdk/pull/1178
Fix conversation deletion error handling for permission issues by @rbren in https://github.com/OpenHands/software-agent-sdk/pull/967
Fix for broken build tests by @tofarr in https://github.com/OpenHands/software-agent-sdk/pull/1181
feat: add configurable shell path for Terminal tool by @mr-karan in https://github.com/OpenHands/software-agent-sdk/pull/1146
only include HTTP exc_info when DEBUG mode is on by @xingyaoww in https://github.com/OpenHands/software-agent-sdk/pull/1184
chore(workflow): restrict auto-assigned reviewers to maintainers (write+) by @enyst in https://github.com/OpenHands/software-agent-sdk/pull/1176
Port over Critic system from benchmark project by @xingyaoww in https://github.com/OpenHands/software-agent-sdk/pull/1171
Fix for issue where LLM API Key is saved as null by @tofarr in https://github.com/OpenHands/software-agent-sdk/pull/1180
Integrate deprecation for systematic deprecation management by @ryanhoangt in https://github.com/OpenHands/software-agent-sdk/pull/1155
Fix for issue where cipher was not set on deserialize by @tofarr in https://github.com/OpenHands/software-agent-sdk/pull/1187
Support Gemini-3 by @xingyaoww in https://github.com/OpenHands/software-agent-sdk/pull/1190

New Contributors

@reecursion made their first contribution in https://github.com/OpenHands/software-agent-sdk/pull/1175
@LuneZ99 made their first contribution in https://github.com/OpenHands/software-agent-sdk/pull/1178
@mr-karan made their first contribution in https://github.com/OpenHands/software-agent-sdk/pull/1146

Full Changelog: https://github.com/OpenHands/software-agent-sdk/compare/v1.1.0...v1.2.0

Open source →

1.1.011 Nov 2025

Added support for glm 4.6 and kimi-k2 (including *extended thinking* mode).

Highlights

Added support for glm 4.6 and kimi-k2 (including extended thinking mode).
Introduced TokenEvent for RL and a new ConversationErrorEvent.
Improved visualization of CondensationRequest
Telemetry improvements: automatic log directory creation
Observability: ad support for OpenTelemetry via Laminar.
FileEditor now supports reading image file.

What's Changed

Added

Support glm 4.6 by @simonrosenberg in https://github.com/OpenHands/software-agent-sdk/pull/1046
Support kimi-k2-thinking by @ryanhoangt in https://github.com/OpenHands/software-agent-sdk/pull/1093
Add image file viewing support to FileEditor by @SmartManoj in https://github.com/OpenHands/software-agent-sdk/pull/1016
Support kimi-k2 extended thinking, fix prompt caching stats, fix max output by @xingyaoww in https://github.com/OpenHands/software-agent-sdk/pull/1133
examples: add observability example by @enyst in https://github.com/OpenHands/software-agent-sdk/pull/1131
add TokenEvent and TokenEvent processing for RL by @lintangsutawika in https://github.com/OpenHands/software-agent-sdk/pull/1123
visualizer: render CondensationRequest nicely by @enyst in https://github.com/OpenHands/software-agent-sdk/pull/1113
Auto-create log directory for telemetry by @SmartManoj in https://github.com/OpenHands/software-agent-sdk/pull/1111

Fixes and Improvements

Fix Datadog log-query to use Logs API instead of Error Tracking API by @neubig in https://github.com/OpenHands/software-agent-sdk/pull/1043
fix: recursively parse nested tools params by @Edward-sy in https://github.com/OpenHands/software-agent-sdk/pull/1011
refactor: git SHA-based Docker tags by default, versioned tags only on releases by @xingyaoww in https://github.com/OpenHands/software-agent-sdk/pull/1088
fix race during cleanup by @enyst in https://github.com/OpenHands/software-agent-sdk/pull/1095
Revert "Add FallbackRouter for LLM failover support" by @xingyaoww in https://github.com/OpenHands/software-agent-sdk/pull/1105
Fix Remote Runtime API Authentication by @xingyaoww in https://github.com/OpenHands/software-agent-sdk/pull/1090
We now have an explicit ConversationErrorEvent in the stream by @tofarr in https://github.com/OpenHands/software-agent-sdk/pull/1106
Remove default path from ENV_ROOT assignment by @SmartManoj in https://github.com/OpenHands/software-agent-sdk/pull/1050
Use shell:bash in run-examples workflow by @xingyaoww in https://github.com/OpenHands/software-agent-sdk/pull/1057
Allow override git ref/sha in docker build by configuring BuildConfig by @xingyaoww in https://github.com/OpenHands/software-agent-sdk/pull/1100
Fix: LLM telemetry logging by @enyst in https://github.com/OpenHands/software-agent-sdk/pull/1028
fix(condenser): When condensation is triggered by the unhandled_condensation_request condition, it will result in empty condensation. by @CLFutureX in https://github.com/OpenHands/software-agent-sdk/pull/1034
Enable Ruff checks for mutable defaults (B006, B008, B039, RUF012) by @enyst in https://github.com/OpenHands/software-agent-sdk/pull/1055
Handle nested field schema formatting for tool description by @ryanhoangt in https://github.com/OpenHands/software-agent-sdk/pull/1001
Fix resume reconciliation by treating litellm_extra_body as runtime override by @enyst in https://github.com/OpenHands/software-agent-sdk/pull/1115
tests(llm): add explicit retry behavior tests for empty choices by @enyst in https://github.com/OpenHands/software-agent-sdk/pull/1107
SDK: export ConversationExecutionStatus via public API by @enyst in https://github.com/OpenHands/software-agent-sdk/pull/1126
fix(backend): unable to create a new conversation using the planning agent’s tools. by @hieptl in https://github.com/OpenHands/software-agent-sdk/pull/1125
Simplify API Workspace example with image_pull_policy argument; fix erroneous clean up of client by @xingyaoww in https://github.com/OpenHands/software-agent-sdk/pull/1127
refactor: move TokenEvent from llm_convertible to event by @xingyaoww in https://github.com/OpenHands/software-agent-sdk/pull/1130
Fix success rate calculation to exclude skipped tests by @xingyaoww in https://github.com/OpenHands/software-agent-sdk/pull/1136
Refactor: Always include risk fields by @malhotra5 in https://github.com/OpenHands/software-agent-sdk/pull/1052

New Contributors

@Edward-sy made their first contribution in https://github.com/OpenHands/software-agent-sdk/pull/1011
@lintangsutawika made their first contribution in https://github.com/OpenHands/software-agent-sdk/pull/1123

Full Changelog: https://github.com/OpenHands/software-agent-sdk/compare/1.0.0...v1.1.0

Open source →

1.0.06 Nov 2025Security

Fix VNC security vulnerability by defaulting to disabled by @rbren in https://github.com/OpenHands/software-agent-sdk/pull/974

Official V1 SDK Release 🎉

Documentation: https://docs.openhands.dev/
Evaluation: https://docs.google.com/spreadsheets/d/1wOUdFCMyY6Nt0AIqF705KN4JKOWgeI4wUGUP60krXXs/edit?gid=811504672#gid=811504672
Paper: https://arxiv.org/abs/2511.03690

What's Changed

Agent Server UUIDs without hyphens by @tofarr in https://github.com/OpenHands/software-agent-sdk/pull/960
Enhance README with branding and resources by @xingyaoww in https://github.com/OpenHands/software-agent-sdk/pull/961
Default to function calling enabled, remove FUNCTION_CALLING_PATTERNS allowlist by @xingyaoww in https://github.com/OpenHands/software-agent-sdk/pull/956
Rename SecretsManager to SecretRegistry by @enyst in https://github.com/OpenHands/software-agent-sdk/pull/965
Rename ConversationState field to secret_registry with read-alias by @enyst in https://github.com/OpenHands/software-agent-sdk/pull/969
Add codex-mini-latest to RESPONSES_API_PATTERNS by @YakshithK in https://github.com/OpenHands/software-agent-sdk/pull/934
Fix double dollar sign in visualizer metrics display by @xingyaoww in https://github.com/OpenHands/software-agent-sdk/pull/938
fix browser tool cleanup by @dinmukhamedm in https://github.com/OpenHands/software-agent-sdk/pull/942
feat: Add llm_response_id to all LLM generated events by @ApGa in https://github.com/OpenHands/software-agent-sdk/pull/930
Fix formatting: remove extra blank line in model_features.py by @xingyaoww in https://github.com/OpenHands/software-agent-sdk/pull/973
Fix VNC security vulnerability by defaulting to disabled by @rbren in https://github.com/OpenHands/software-agent-sdk/pull/974
Fix wrong server path for file upload endpoint by @ryanhoangt in https://github.com/OpenHands/software-agent-sdk/pull/951
Fix bash executor session reuse by resetting closed sessions by @softpudding in https://github.com/OpenHands/software-agent-sdk/pull/964
Fix duplicate reviewer assignment in assign-reviews workflow by @neubig in https://github.com/OpenHands/software-agent-sdk/pull/827
Add logging for file upload and download actions by @rbren in https://github.com/OpenHands/software-agent-sdk/pull/993
sdk(llm): Stop model-name whack-a-mole: revert to core family substring matching by @enyst in https://github.com/OpenHands/software-agent-sdk/pull/879
Add glm4.6 to FORCE_STRING_SERIALIZER by @simonrosenberg in https://github.com/OpenHands/software-agent-sdk/pull/994
update patterns in FORCE_STRING_SERIALIZER_PATTERNS by @simonrosenberg in https://github.com/OpenHands/software-agent-sdk/pull/995
Add tests for string serializer by @enyst in https://github.com/OpenHands/software-agent-sdk/pull/996
Add cost reporting for run-examples.yml and schedule scripts to run daily by @ryanhoangt in https://github.com/OpenHands/software-agent-sdk/pull/977
Remove default MCP servers from preset agents by @xingyaoww in https://github.com/OpenHands/software-agent-sdk/pull/984
Bug: fix Visualizer "Message To User" that should be "Message from User" by @simonrosenberg in https://github.com/OpenHands/software-agent-sdk/pull/1002
Tweak review workflow for draft PRs by @enyst in https://github.com/OpenHands/software-agent-sdk/pull/997
Fix(llm):When LLM calls are retried, it can lead to inaccurate records. by @CLFutureX in https://github.com/OpenHands/software-agent-sdk/pull/949
Normalize SDK errors for clients by @enyst in https://github.com/OpenHands/software-agent-sdk/pull/980
Add Datadog debugging example with API integration tools by @neubig in https://github.com/OpenHands/software-agent-sdk/pull/272
Add Sphinx-compatible docstrings to core SDK classes by @rbren in https://github.com/OpenHands/software-agent-sdk/pull/1006
Refactor ToolDefinition architecture to use subclass pattern for all tools by @xingyaoww in https://github.com/OpenHands/software-agent-sdk/pull/971
Fix datadog debugging workflow example by @neubig in https://github.com/OpenHands/software-agent-sdk/pull/1009
feat(backend): enhance search_conversation_events functions with timestamp filtering support by @hieptl in https://github.com/OpenHands/software-agent-sdk/pull/695
Add comprehensive unit tests for conversation_router.py endpoints by @tofarr in https://github.com/OpenHands/software-agent-sdk/pull/1010
Update example API key in README by @SmartManoj in https://github.com/OpenHands/software-agent-sdk/pull/1013
Migrate to Blacksmith official Docker actions with native multi-platform builds by @xingyaoww in https://github.com/OpenHands/software-agent-sdk/pull/990
Fix/remove docstring for OpenHands tool package by @rbren in https://github.com/OpenHands/software-agent-sdk/pull/1018
feat(llm): forward extra_headers to LiteLLM completion and responses by @enyst in https://github.com/OpenHands/software-agent-sdk/pull/733
refactor: rename AgentExecutionStatus -> ConversationExecutionStatus & state's agent_status -> executiong_status by @simonrosenberg in https://github.com/OpenHands/software-agent-sdk/pull/839
Add OpenTelemetry with Laminar SDK by @dinmukhamedm in https://github.com/OpenHands/software-agent-sdk/pull/681
More agent_status to execution_status_updates, and error state by @tofarr in https://github.com/OpenHands/software-agent-sdk/pull/1019
Recognize GPT‑5 context overflow and trigger condenser (Fixes #776) by @enyst in https://github.com/OpenHands/software-agent-sdk/pull/783
Rename metadata field to litellm_extra_body and add custom config support by @li-boxuan in https://github.com/OpenHands/software-agent-sdk/pull/837
Fix error in 422 logs due to blank bearer token by @tofarr in https://github.com/OpenHands/software-agent-sdk/pull/1022
Fix abbreviation formatting in visualizer by @SmartManoj in https://github.com/OpenHands/software-agent-sdk/pull/989
Implement automatic tool registration on import by @malhotra5 in https://github.com/OpenHands/software-agent-sdk/pull/862
feat: Add automatic loading of user skills from home directory by @whywaita in https://github.com/OpenHands/software-agent-sdk/pull/950
Fix multi-arch manifest merge to use correct tag names by @xingyaoww in https://github.com/OpenHands/software-agent-sdk/pull/1024
Disable Copilot chat command center icon by default (parity with OpenHands PR #11589) by @enyst in https://github.com/OpenHands/software-agent-sdk/pull/1032
Fix Docker cache tag length exceeding 128 character limit by @xingyaoww in https://github.com/OpenHands/software-agent-sdk/pull/1029
Fix Laminar span stack warning in LocalConversation by @jpshackelford in https://github.com/OpenHands/software-agent-sdk/pull/1039
Include 03_browser_use_with_docker_sandboxed_server.py in test-examples workflow by @xingyaoww in https://github.com/OpenHands/software-agent-sdk/pull/1021
A couple of fixes for OpenAI o3 by @enyst in https://github.com/OpenHands/software-agent-sdk/pull/1031
refactor: standardize Observation base class by @simonrosenberg in https://github.com/OpenHands/software-agent-sdk/pull/929
Fix: Make reasoning_effort optional by @Chesars in https://github.com/OpenHands/software-agent-sdk/pull/1004
Simplify hello world example by @rbren in https://github.com/OpenHands/software-agent-sdk/pull/1047
Add reviewer trigger for PR review workflow by @simonrosenberg in https://github.com/OpenHands/software-agent-sdk/pull/1037
Fix RemoteConversation stats field mismatch and state updates by @xingyaoww in https://github.com/OpenHands/software-agent-sdk/pull/1042
Rename execute_bash to terminal & rename BashTool to TerminalTool by @SmartManoj in https://github.com/OpenHands/software-agent-sdk/pull/1033
Improve visualization API: allow passing ConversationVisualizer directly by @xingyaoww in https://github.com/OpenHands/software-agent-sdk/pull/1025
v1.0.0 release by @xingyaoww in https://github.com/OpenHands/software-agent-sdk/pull/1056

New Contributors

@YakshithK made their first contribution in https://github.com/OpenHands/software-agent-sdk/pull/934
@ApGa made their first contribution in https://github.com/OpenHands/software-agent-sdk/pull/930
@softpudding made their first contribution in https://github.com/OpenHands/software-agent-sdk/pull/964
@CLFutureX made their first contribution in https://github.com/OpenHands/software-agent-sdk/pull/949
@SmartManoj made their first contribution in https://github.com/OpenHands/software-agent-sdk/pull/1013
@whywaita made their first contribution in https://github.com/OpenHands/software-agent-sdk/pull/950
@Chesars made their first contribution in https://github.com/OpenHands/software-agent-sdk/pull/1004

Full Changelog: https://github.com/OpenHands/software-agent-sdk/compare/1.0.0a5...1.0.0

Open source →

1.0.0a65 Nov 2025pre-release

Nothing published for this version

1.0.0a529 Oct 2025pre-release

Fix PR review workflow subdirectory paths by @simonrosenberg in https://github.com/OpenHands/software-agent-sdk/pull/824

What's Changed

Fix PR review workflow subdirectory paths by @simonrosenberg in https://github.com/OpenHands/software-agent-sdk/pull/824
Fix httpx.UnsupportedProtocol error by setting base_url on clients by @neubig in https://github.com/OpenHands/software-agent-sdk/pull/801
docs: set tags=["Server Details"] for / path by @xingyaoww in https://github.com/OpenHands/software-agent-sdk/pull/829
Enforce batch atomicity for condenser by @ryanhoangt in https://github.com/OpenHands/software-agent-sdk/pull/775
Replace All-Hands-AI references with OpenHands by @rbren in https://github.com/OpenHands/software-agent-sdk/pull/830
Fix: Make adding security_analyzer optional for get_default_agent by @jimwhite in https://github.com/OpenHands/software-agent-sdk/pull/802
Add test to reproduce bug #825: execute_command includes previous output by @neubig in https://github.com/OpenHands/software-agent-sdk/pull/828
Add FunctionCallValidationError handling & claude-haiku-4-5 to function calling supported models by @ryanhoangt in https://github.com/OpenHands/software-agent-sdk/pull/832
Set add_security_analyzer=False by default and simplify examples by @xingyaoww in https://github.com/OpenHands/software-agent-sdk/pull/846
Debug: glob tool fails when pattern contains path by @simonrosenberg in https://github.com/OpenHands/software-agent-sdk/pull/718
.github/workflows: Migrate workflows to Blacksmith runners by @blacksmith-sh[bot] in https://github.com/OpenHands/software-agent-sdk/pull/831
Rename "microagents" to "skills" throughout the repository by @neubig in https://github.com/OpenHands/software-agent-sdk/pull/820
Fix FastAPI path parameters to handle file paths with forward slashes by @amanape in https://github.com/OpenHands/software-agent-sdk/pull/848
Env template generator by @tofarr in https://github.com/OpenHands/software-agent-sdk/pull/847
Socket logging by @tofarr in https://github.com/OpenHands/software-agent-sdk/pull/858
Simplify examples for docs by @xingyaoww in https://github.com/OpenHands/software-agent-sdk/pull/859
fix: ReadTimeout when calling 01_convo_with_local_agent_server.py by @juanmichelini in https://github.com/OpenHands/software-agent-sdk/pull/857
Simplify remote agent server examples by @xingyaoww in https://github.com/OpenHands/software-agent-sdk/pull/868
Add conversation to tool calls by @simonrosenberg in https://github.com/OpenHands/software-agent-sdk/pull/877
Add workflow to check documented examples by @xingyaoww in https://github.com/OpenHands/software-agent-sdk/pull/878
chore(examples): env-driven endpoints, platform detection; update domains and API var names by @enyst in https://github.com/OpenHands/software-agent-sdk/pull/872
Fix DockerWorkspace execute_command timeout issue by @neubig in https://github.com/OpenHands/software-agent-sdk/pull/871
tests(remote): silence pytest collection warning by renaming helper class by @enyst in https://github.com/OpenHands/software-agent-sdk/pull/880
Fix agent status not transitioning to running status by @ryanhoangt in https://github.com/OpenHands/software-agent-sdk/pull/884
Dependency Injection for EventService by @tofarr in https://github.com/OpenHands/software-agent-sdk/pull/885
add asssign reviews workflow to examples by @simonrosenberg in https://github.com/OpenHands/software-agent-sdk/pull/883
feat(tests): Add comprehensive test suite for RemoteConversation implementation by @xingyaoww in https://github.com/OpenHands/software-agent-sdk/pull/451
Switch back to pyright + stricter typing to allow vscode to resolve imports by @xingyaoww in https://github.com/OpenHands/software-agent-sdk/pull/869
Fix pyright issues in agent status transition tests by @enyst in https://github.com/OpenHands/software-agent-sdk/pull/895
Fix MCP tool validation error by excluding 'kind' field from action data by @xingyaoww in https://github.com/OpenHands/software-agent-sdk/pull/887
Add regression test for bash command endpoint fix (#866) by @xingyaoww in https://github.com/OpenHands/software-agent-sdk/pull/881
Update sdk version to 1.0.0a4 by @xingyaoww in https://github.com/OpenHands/software-agent-sdk/pull/901
Fix PydanticSerializationUnexpectedValue warnings in log_completions by @xingyaoww in https://github.com/OpenHands/software-agent-sdk/pull/904
Fix #905: Handle None context_window for unmapped models by @xingyaoww in https://github.com/OpenHands/software-agent-sdk/pull/906
Add workflow to run examples script & fix failed integration tests by @ryanhoangt in https://github.com/OpenHands/software-agent-sdk/pull/914
AsyncRemoteWorkspace now has working directory by @tofarr in https://github.com/OpenHands/software-agent-sdk/pull/888
refactor: move agent_final_response to utility function with proper type checking by @xingyaoww in https://github.com/OpenHands/software-agent-sdk/pull/919
Feat encrypted secrets by @tofarr in https://github.com/OpenHands/software-agent-sdk/pull/893
chore: remove Dependabot config (uv workspace unsupported) by @enyst in https://github.com/OpenHands/software-agent-sdk/pull/897
Bump the version-all group with 3 updates by @dependabot[bot] in https://github.com/OpenHands/software-agent-sdk/pull/913
Add docs workflow to repo.md and fix pyright type checking by @xingyaoww in https://github.com/OpenHands/software-agent-sdk/pull/924
feat: Implement blocking delegation by @simonrosenberg in https://github.com/OpenHands/software-agent-sdk/pull/908
refactor: Update ToolExecutor interface to use LocalConversation instead of BaseConversation by @xingyaoww in https://github.com/OpenHands/software-agent-sdk/pull/925
Fix #916: Suppress Pydantic serializer warnings from litellm by @xingyaoww in https://github.com/OpenHands/software-agent-sdk/pull/917
Update README to refer to documentation site by @xingyaoww in https://github.com/OpenHands/software-agent-sdk/pull/928
Feat git operations by @tofarr in https://github.com/OpenHands/software-agent-sdk/pull/863
refactor: Clean up _configure_bash_tools_env_provider anti-pattern by @xingyaoww in https://github.com/OpenHands/software-agent-sdk/pull/922
Update integration test to use OpenHands proxy instead of litellm proxy by @simonrosenberg in https://github.com/OpenHands/software-agent-sdk/pull/944
feat: Post integration test results to tracker issue for daily runs by @simonrosenberg in https://github.com/OpenHands/software-agent-sdk/pull/940
Update litellm proxy path in workflows by @simonrosenberg in https://github.com/OpenHands/software-agent-sdk/pull/946
bump browser-use version, infer screenshot mime type dynamically by @dinmukhamedm in https://github.com/OpenHands/software-agent-sdk/pull/935
change base url in integration tests by @simonrosenberg in https://github.com/OpenHands/software-agent-sdk/pull/947
feat: support absolute paths for system_prompt_filename by @xingyaoww in https://github.com/OpenHands/software-agent-sdk/pull/902
Fix integration tests to work with fork PRs by @xingyaoww in https://github.com/OpenHands/software-agent-sdk/pull/932
refactor: fix naming inconsistency by @simonrosenberg in https://github.com/OpenHands/software-agent-sdk/pull/958
release version 1.0.0a5 by @xingyaoww in https://github.com/OpenHands/software-agent-sdk/pull/959

New Contributors

@jimwhite made their first contribution in https://github.com/OpenHands/software-agent-sdk/pull/802
@blacksmith-sh[bot] made their first contribution in https://github.com/OpenHands/software-agent-sdk/pull/831
@amanape made their first contribution in https://github.com/OpenHands/software-agent-sdk/pull/848
@dinmukhamedm made their first contribution in https://github.com/OpenHands/software-agent-sdk/pull/935

Full Changelog: https://github.com/OpenHands/software-agent-sdk/compare/1.0.0a3...1.0.0a5

Open source →

1.0.0a425 Oct 2025pre-release

Nothing published for this version

1.0.0a320 Oct 2025pre-releaseDeprecated

Fix websocket deprecation warning by using wsproto by @xingyaoww in https://github.com/All-Hands-AI/agent-sdk/pull/508

What's Changed

Add example for automated TODO management with GitHub Actions by @simonrosenberg in https://github.com/All-Hands-AI/agent-sdk/pull/758
Rename LLM service identifiers to usage_id by @enyst in https://github.com/All-Hands-AI/agent-sdk/pull/799
Bump the version-all group with 4 updates by @dependabot[bot] in https://github.com/All-Hands-AI/agent-sdk/pull/810
Fix assign-reviews workflow subdirectory paths by @neubig in https://github.com/All-Hands-AI/agent-sdk/pull/807
Fix websocket deprecation warning by using wsproto by @xingyaoww in https://github.com/All-Hands-AI/agent-sdk/pull/508
Fix: Make Check OpenAPI Schema workflow work with fork PRs by @xingyaoww in https://github.com/All-Hands-AI/agent-sdk/pull/814
Setting VSCode token to match SESSION_API_KEY. by @tofarr in https://github.com/All-Hands-AI/agent-sdk/pull/793
Fix deprecated @model_validator usage in LLMSummarizingCondenser by @xingyaoww in https://github.com/All-Hands-AI/agent-sdk/pull/817
Fix: Skip Docker build and push for fork PRs by @xingyaoww in https://github.com/All-Hands-AI/agent-sdk/pull/821
1.0.0a3 pre-release by @xingyaoww in https://github.com/All-Hands-AI/agent-sdk/pull/822

Full Changelog: https://github.com/All-Hands-AI/agent-sdk/compare/1.0.0a2...1.0.0a3

Open source →

1.0.0a217 Oct 2025pre-release

Sync readme to changes by @enyst in https://github.com/All-Hands-AI/agent-sdk/pull/734

What's Changed

Sync readme to changes by @enyst in https://github.com/All-Hands-AI/agent-sdk/pull/734
use uv-managed python for "source" docker by @xingyaoww in https://github.com/All-Hands-AI/agent-sdk/pull/737
Render user rejections in visualizer by @enyst in https://github.com/All-Hands-AI/agent-sdk/pull/746
fix: make ThinkAction skip user confirmation like FinishAction by @malhotra5 in https://github.com/All-Hands-AI/agent-sdk/pull/742
Add GitHub Actions workflow for scheduled maintenance tasks by @neubig in https://github.com/All-Hands-AI/agent-sdk/pull/672
Enable unused variable detection in pre-commit hooks by @xingyaoww in https://github.com/All-Hands-AI/agent-sdk/pull/726
Add optional conversation_id parameter to StartConversationRequest by @malhotra5 in https://github.com/All-Hands-AI/agent-sdk/pull/741
Add AsyncRemoteWorkspace and sync with APIRemoteWorkspace by @tofarr in https://github.com/All-Hands-AI/agent-sdk/pull/735
Update all SDK examples to read LLM_MODEL and LLM_BASE_URL from environment by @neubig in https://github.com/All-Hands-AI/agent-sdk/pull/740
Add PR review workflow for fine-grained review comments by @simonrosenberg in https://github.com/All-Hands-AI/agent-sdk/pull/751
Add "assign reviewers" workflow by @neubig in https://github.com/All-Hands-AI/agent-sdk/pull/756
Update assign reviewers script based on experience by @neubig in https://github.com/All-Hands-AI/agent-sdk/pull/760
Fix for URL endpoint location by @tofarr in https://github.com/All-Hands-AI/agent-sdk/pull/503
Bump the version-all group with 2 updates by @dependabot[bot] in https://github.com/All-Hands-AI/agent-sdk/pull/535
refactor: remove duplicated example and rename folder by @xingyaoww in https://github.com/All-Hands-AI/agent-sdk/pull/765
Extend context truncation cases: broaden vLLM error detection by @enyst in https://github.com/All-Hands-AI/agent-sdk/pull/759
Add support for claude-4-5-haiku by @ryanhoangt in https://github.com/All-Hands-AI/agent-sdk/pull/768
Fix PyPI Wheel by @xingyaoww in https://github.com/All-Hands-AI/agent-sdk/pull/752
Print conversation ID when an exception stops the run (fixes #777) by @enyst in https://github.com/All-Hands-AI/agent-sdk/pull/779
Feat: Add Configurable Plan Template to Planning Agent by @simonrosenberg in https://github.com/All-Hands-AI/agent-sdk/pull/778
refactor(llm): extract kwargs normalization into pure options selectors by @enyst in https://github.com/All-Hands-AI/agent-sdk/pull/643
Add basedpyright to pre-commit and fix all type errors by @xingyaoww in https://github.com/All-Hands-AI/agent-sdk/pull/593
fix: deploy changes to docs repo by @xingyaoww in https://github.com/All-Hands-AI/agent-sdk/pull/786
update deploy-to-docs workflow by @xingyaoww in https://github.com/All-Hands-AI/agent-sdk/pull/788

Full Changelog: https://github.com/All-Hands-AI/agent-sdk/compare/1.0.0a1...1.0.0a2

Open source →

1.0.0a114 Oct 2025pre-releaseDeprecated

Remove deprecated params={'working_dir'} and params={'workspace_root'} patterns from tools by @xingyaoww in https://github.com/All-Hands-AI/agent-sdk/…

What's Changed

Setup Makefile, Pre-commit, and initial Tool and Schema definition by @xingyaoww in https://github.com/All-Hands-AI/agent-sdk/pull/1
Port over file editor tool by @xingyaoww in https://github.com/All-Hands-AI/agent-sdk/pull/2
Fix tool schema and add relevant tests; Add bash implmentations by @xingyaoww in https://github.com/All-Hands-AI/agent-sdk/pull/4
Update README and add repo.md by @xingyaoww in https://github.com/All-Hands-AI/agent-sdk/pull/5
ci: add coverage report and summary to tests workflow (closes #3) by @xingyaoww in https://github.com/All-Hands-AI/agent-sdk/pull/6
Port over BashSession and tests by @xingyaoww in https://github.com/All-Hands-AI/agent-sdk/pull/7
Openhands/move coverage badge to docs assets by @xingyaoww in https://github.com/All-Hands-AI/agent-sdk/pull/10
Update ci badge on pr only by @xingyaoww in https://github.com/All-Hands-AI/agent-sdk/pull/11
Add minimal config & port over the LLM class by @xingyaoww in https://github.com/All-Hands-AI/agent-sdk/pull/12
Only trigger tests on PR by @xingyaoww in https://github.com/All-Hands-AI/agent-sdk/pull/15
Port over minimal set of agent, config, context, microagent; adjust existing tool definition by @xingyaoww in https://github.com/All-Hands-AI/agent-sdk/pull/18
Improve tool call visualization by @xingyaoww in https://github.com/All-Hands-AI/agent-sdk/pull/20
run test on main by @xingyaoww in https://github.com/All-Hands-AI/agent-sdk/pull/21
logger: set LiteLLM loggers to WARNING by default; enable DEBUG with DEBUG_LLM confirmation by @xingyaoww in https://github.com/All-Hands-AI/agent-sdk/pull/23
Remove json repair by @xingyaoww in https://github.com/All-Hands-AI/agent-sdk/pull/24
Add ruff configuration for import sorting by @xingyaoww in https://github.com/All-Hands-AI/agent-sdk/pull/17
Simplify pre-commit CI by @xingyaoww in https://github.com/All-Hands-AI/agent-sdk/pull/26
rename package openhands-ai to openhands-sdk by @xingyaoww in https://github.com/All-Hands-AI/agent-sdk/pull/27
Rename openhands-ai as openhands-sdk by @xingyaoww in https://github.com/All-Hands-AI/agent-sdk/pull/28
Rename unnecessary test_ruff_config.py by @xingyaoww in https://github.com/All-Hands-AI/agent-sdk/pull/29
Restructure agent-sdk repository into UV workspace with simplified structure by @xingyaoww in https://github.com/All-Hands-AI/agent-sdk/pull/33
add openhands setup.sh by @xingyaoww in https://github.com/All-Hands-AI/agent-sdk/pull/31
ci: make coverage PR comment sticky to avoid duplicates (fixes #34) by @xingyaoww in https://github.com/All-Hands-AI/agent-sdk/pull/35
refactor repo by creating context/; move looping logic into conversation.py by @xingyaoww in https://github.com/All-Hands-AI/agent-sdk/pull/36
Update repo.md for tests by @xingyaoww in https://github.com/All-Hands-AI/agent-sdk/pull/37
support multiple callback fn; improve examples of how to add callback from script by @xingyaoww in https://github.com/All-Hands-AI/agent-sdk/pull/38
refactor FinishTool to builtin tools for agents by @xingyaoww in https://github.com/All-Hands-AI/agent-sdk/pull/39
follow torvards-mode suggestion: integrate lock with state, and modify state in-place by @xingyaoww in https://github.com/All-Hands-AI/agent-sdk/pull/40
Make unit test jobs conditional based on folder changes by @xingyaoww in https://github.com/All-Hands-AI/agent-sdk/pull/46
refactor the system to use Event Types by @xingyaoww in https://github.com/All-Hands-AI/agent-sdk/pull/44
Conversation: make state.events.append a default callback; remove manual appends in CodeActAgent by @xingyaoww in https://github.com/All-Hands-AI/agent-sdk/pull/57
Remove prompt manager and instead using simpler jinja util function by @xingyaoww in https://github.com/All-Hands-AI/agent-sdk/pull/63
Port over microagent changes in https://github.com/All-Hands-AI/OpenHands/pull/10528 by @xingyaoww in https://github.com/All-Hands-AI/agent-sdk/pull/65
Add ruff format to pre-commit hooks for consistent code formatting by @xingyaoww in https://github.com/All-Hands-AI/agent-sdk/pull/67
Fix: Include Jinja2 template files in package distribution by @malhotra5 in https://github.com/All-Hands-AI/agent-sdk/pull/69
Reduce line-length from 500 to 88 and fix pre-commit errors by @xingyaoww in https://github.com/All-Hands-AI/agent-sdk/pull/68
Add remaining LLM fixes and LLM registry by @ryanhoangt in https://github.com/All-Hands-AI/agent-sdk/pull/70
Port LLM tests from OpenHands to agent-sdk by @enyst in https://github.com/All-Hands-AI/agent-sdk/pull/48
Standardize project test structure by @xingyaoww in https://github.com/All-Hands-AI/agent-sdk/pull/75
Standardize on absolute imports (PEP 8) and ensure repo package precedence by @xingyaoww in https://github.com/All-Hands-AI/agent-sdk/pull/74
Rename CodeActAgent to Agent and update folder structure by @xingyaoww in https://github.com/All-Hands-AI/agent-sdk/pull/78
Rename openhands.core to openhands.sdk by @xingyaoww in https://github.com/All-Hands-AI/agent-sdk/pull/73
Core context condensation implementation by @csmith49 in https://github.com/All-Hands-AI/agent-sdk/pull/61
test: Add comprehensive tests for events_to_messages conversions by @xingyaoww in https://github.com/All-Hands-AI/agent-sdk/pull/81
Support microagents and unify it within AgentContext by @xingyaoww in https://github.com/All-Hands-AI/agent-sdk/pull/71
feat: Add BashTool and FileEditorTool subclasses for simplified tool initialization by @xingyaoww in https://github.com/All-Hands-AI/agent-sdk/pull/88
Fix coverage upload issue in CI workflow by @xingyaoww in https://github.com/All-Hands-AI/agent-sdk/pull/90
Optimize pre-commit speed by setting always_run to false by @xingyaoww in https://github.com/All-Hands-AI/agent-sdk/pull/96
fix typo in prompt suffix by @xingyaoww in https://github.com/All-Hands-AI/agent-sdk/pull/97
Configure Dependabot for uv and GitHub Actions by @xingyaoww in https://github.com/All-Hands-AI/agent-sdk/pull/98
fix: Minor test cleanup by @csmith49 in https://github.com/All-Hands-AI/agent-sdk/pull/99
Bump astral-sh/setup-uv from 3 to 6 by @dependabot[bot] in https://github.com/All-Hands-AI/agent-sdk/pull/100
Add open-pull-requests-limit and groups to dependabot by @xingyaoww in https://github.com/All-Hands-AI/agent-sdk/pull/105
Bump the version-all group with 4 updates by @dependabot[bot] in https://github.com/All-Hands-AI/agent-sdk/pull/108
Bump actions/checkout from 4 to 5 by @dependabot[bot] in https://github.com/All-Hands-AI/agent-sdk/pull/104
Port config tests from OpenHands to agent-sdk by @enyst in https://github.com/All-Hands-AI/agent-sdk/pull/49
fix: Idempotent Message serialization by @csmith49 in https://github.com/All-Hands-AI/agent-sdk/pull/106
Bump actions/download-artifact from 4 to 5 by @dependabot[bot] in https://github.com/All-Hands-AI/agent-sdk/pull/101
Implement confirmation mode for OpenHands agent SDK by @malhotra5 in https://github.com/All-Hands-AI/agent-sdk/pull/80
feat(agent): simplify security guidelines in system prompt by @enyst in https://github.com/All-Hands-AI/agent-sdk/pull/115
Update system prompt SECURITY section to match OpenHands PR #10822 by @xingyaoww in https://github.com/All-Hands-AI/agent-sdk/pull/114
feat: improve visualizer by @xingyaoww in https://github.com/All-Hands-AI/agent-sdk/pull/119
Migrate CLI subprocess shell and refactor BashSession architecture by @xingyaoww in https://github.com/All-Hands-AI/agent-sdk/pull/109
Refactor LLMConfig into LLM pydantic class & large refactor of LLM class logic by @xingyaoww in https://github.com/All-Hands-AI/agent-sdk/pull/91
Clean up all examples; add interactive bash example; improve visualizer for action by @xingyaoww in https://github.com/All-Hands-AI/agent-sdk/pull/122
Update README.md and fix an AgentContext bug by @xingyaoww in https://github.com/All-Hands-AI/agent-sdk/pull/127
Port runtime bash tests by @ryanhoangt in https://github.com/All-Hands-AI/agent-sdk/pull/84
Migrate MCP clients by @ryanhoangt in https://github.com/All-Hands-AI/agent-sdk/pull/94
feat(logger): improve logger using RichHandler by @xingyaoww in https://github.com/All-Hands-AI/agent-sdk/pull/132
Fix confusing LLMRegistry API after LLM/LLMConfig merge by @xingyaoww in https://github.com/All-Hands-AI/agent-sdk/pull/129
Add pause/resume functionality to agent-sdk conversation system by @malhotra5 in https://github.com/All-Hands-AI/agent-sdk/pull/118
Add truncation to ExecuteBashObservation and TextContent with warning checks by @xingyaoww in https://github.com/All-Hands-AI/agent-sdk/pull/124
docs: Add comprehensive tools system documentation by @xingyaoww in https://github.com/All-Hands-AI/agent-sdk/pull/136
Move tests to a new directory by @xingyaoww in https://github.com/All-Hands-AI/agent-sdk/pull/140
Fix pause event type by @malhotra5 in https://github.com/All-Hands-AI/agent-sdk/pull/144
chore(docs): create docs/ folder and move existing documentation by @enyst in https://github.com/All-Hands-AI/agent-sdk/pull/117
fix: resolve empty API keys to None and add Bedrock model support by @enyst in https://github.com/All-Hands-AI/agent-sdk/pull/141
Fix prompt caching and improve input/output metrics visualization by @xingyaoww in https://github.com/All-Hands-AI/agent-sdk/pull/143
Stabilize tests: telemetry logging warning and Bash PS2 prompt capture by @enyst in https://github.com/All-Hands-AI/agent-sdk/pull/145
Fix issue #10729: Add xai/grok-code-fast-1 to MODELS_WITHOUT_STOP_WORDS by @enyst in https://github.com/All-Hands-AI/agent-sdk/pull/147
Update microagent prompt to be more collaborative and less abrasive by @xingyaoww in https://github.com/All-Hands-AI/agent-sdk/pull/149
feat: Lazy discriminated unions for event/action/observation serialization by @csmith49 in https://github.com/All-Hands-AI/agent-sdk/pull/133
Support microagent repo.md with new MCP tool format by @xingyaoww in https://github.com/All-Hands-AI/agent-sdk/pull/154
Add conversation ID to Conversation class by @xingyaoww in https://github.com/All-Hands-AI/agent-sdk/pull/158
fix(MCP): do not include base action attributes like kind safety_risk for executing MCP Action by @xingyaoww in https://github.com/All-Hands-AI/agent-sdk/pull/156
fix parent fields to include kind by considering model_fields_computed by @xingyaoww in https://github.com/All-Hands-AI/agent-sdk/pull/159
increase mcp timeout to 300 sec by @xingyaoww in https://github.com/All-Hands-AI/agent-sdk/pull/160
Fix switching confirmation mode and approving actions in one step by @malhotra5 in https://github.com/All-Hands-AI/agent-sdk/pull/163
feat: Support reasoning content in Agent SDK by @enyst in https://github.com/All-Hands-AI/agent-sdk/pull/139
Add think and task_tracker tools by @ryanhoangt in https://github.com/All-Hands-AI/agent-sdk/pull/148
Add comprehensive tests for file editor by @ryanhoangt in https://github.com/All-Hands-AI/agent-sdk/pull/165
chore: Merge test_editor_comprehensive with test_basic_operations by @xingyaoww in https://github.com/All-Hands-AI/agent-sdk/pull/168
Add docs for MCP and update README by @ryanhoangt in https://github.com/All-Hands-AI/agent-sdk/pull/166
Make Agent & Tool class a BaseModel with frozen=True by @enyst in https://github.com/All-Hands-AI/agent-sdk/pull/134
Refactor visualizer to use modular approach with individual event classes & consistent color by @xingyaoww in https://github.com/All-Hands-AI/agent-sdk/pull/171
Add visualize_diff functionality to StrReplaceEditorObservation by @malhotra5 in https://github.com/All-Hands-AI/agent-sdk/pull/125
refactor test folder - rename integration -> cross by @simonrosenberg in https://github.com/All-Hands-AI/agent-sdk/pull/178
Make both agent and tool subclass of DiscriminatedUnionMixin & serializable by @xingyaoww in https://github.com/All-Hands-AI/agent-sdk/pull/174
Flatten agent folder since there's only one agent now by @xingyaoww in https://github.com/All-Hands-AI/agent-sdk/pull/179
refactor(llm): extract non-native tool-calling into mixin by @enyst in https://github.com/All-Hands-AI/agent-sdk/pull/182
Update stale documentation for OpenHands Tools System by @xingyaoww in https://github.com/All-Hands-AI/agent-sdk/pull/184
Improve visualization of Agent SDK by @xingyaoww in https://github.com/All-Hands-AI/agent-sdk/pull/187
Serialization of Conversation to FileStore by @xingyaoww in https://github.com/All-Hands-AI/agent-sdk/pull/41
Add test for using LLM for non-function call completions by @xingyaoww in https://github.com/All-Hands-AI/agent-sdk/pull/185
Move prompts in context directory by @enyst in https://github.com/All-Hands-AI/agent-sdk/pull/191
Restructure documentation into comprehensive architecture/ folder and add OH doc update workflow by @xingyaoww in https://github.com/All-Hands-AI/agent-sdk/pull/188
docs: tweak header by @xingyaoww in https://github.com/All-Hands-AI/agent-sdk/pull/196
docs: Add context architecture documentation and improve mermaid chart readability by @xingyaoww in https://github.com/All-Hands-AI/agent-sdk/pull/197
Update documentation to reflect current codebase structure by @xingyaoww in https://github.com/All-Hands-AI/agent-sdk/pull/198
refactor: rename/remove 'integration' from cross tests by @simonrosenberg in https://github.com/All-Hands-AI/agent-sdk/pull/202
refactor: remove emoji from security risk policy by @simonrosenberg in https://github.com/All-Hands-AI/agent-sdk/pull/201
feat: add file deletion step in examples by @simonrosenberg in https://github.com/All-Hands-AI/agent-sdk/pull/200
fix: ImportError during Python shutdown in terminal session cleanup by @xingyaoww in https://github.com/All-Hands-AI/agent-sdk/pull/162
feat(viz): do not show finish action observation; increase margin for all by @xingyaoww in https://github.com/All-Hands-AI/agent-sdk/pull/204
feat: support for setting conversation id when start the conversation by @xingyaoww in https://github.com/All-Hands-AI/agent-sdk/pull/203
move docs from this repo to docs repo and disable sync by @xingyaoww in https://github.com/All-Hands-AI/agent-sdk/pull/207
feat: visualize notes in italic for task tracker by @xingyaoww in https://github.com/All-Hands-AI/agent-sdk/pull/210
Use file-backed list for "events" by @xingyaoww in https://github.com/All-Hands-AI/agent-sdk/pull/205
Add AsyncConversationCallback type and AsyncCallbackWrapper class by @tofarr in https://github.com/All-Hands-AI/agent-sdk/pull/217
chore: remove redundant RetryMixin in llm.py by @xingyaoww in https://github.com/All-Hands-AI/agent-sdk/pull/222
fix microagent typing that breaks openapi.json & remove unused by @xingyaoww in https://github.com/All-Hands-AI/agent-sdk/pull/221
Fix README accuracy issues by @enyst in https://github.com/All-Hands-AI/agent-sdk/pull/225
Fix type hints for get_edit_groups to accept optional parameters by @enyst in https://github.com/All-Hands-AI/agent-sdk/pull/229
Fix FileStore ABC inheritance by @enyst in https://github.com/All-Hands-AI/agent-sdk/pull/227
Remove redundant agent diff check logic by @enyst in https://github.com/All-Hands-AI/agent-sdk/pull/231
Fix LocalFileStore.list() to handle file targets gracefully by @enyst in https://github.com/All-Hands-AI/agent-sdk/pull/228
Expand conversation abbreviations to full names by @enyst in https://github.com/All-Hands-AI/agent-sdk/pull/233
fix: prevent data mutation during visualize by @enyst in https://github.com/All-Hands-AI/agent-sdk/pull/226
do not try to force ImageContent be subclass of mcp image content by @xingyaoww in https://github.com/All-Hands-AI/agent-sdk/pull/212
Replace list[TextContent | ImageContent] with Sequence[...] for type compatibility by @xingyaoww in https://github.com/All-Hands-AI/agent-sdk/pull/238
Improve error handling in TaskTracker JSON loading by @enyst in https://github.com/All-Hands-AI/agent-sdk/pull/230
Make all events immutable by adding frozen=True by @xingyaoww in https://github.com/All-Hands-AI/agent-sdk/pull/241
feat: Add secrets manager for secure credential handling by @malhotra5 in https://github.com/All-Hands-AI/agent-sdk/pull/189
Remove all dim text styling from visualizer by @xingyaoww in https://github.com/All-Hands-AI/agent-sdk/pull/244
Feat: mask secrets by @malhotra5 in https://github.com/All-Hands-AI/agent-sdk/pull/248
Fix execute_bash tool to inherit environment variables from parent process by @xingyaoww in https://github.com/All-Hands-AI/agent-sdk/pull/245
fix: prevent path traversal attacks in LocalFileStore by @enyst in https://github.com/All-Hands-AI/agent-sdk/pull/232
Add repomix tool for codebase packaging and analysis by @xingyaoww in https://github.com/All-Hands-AI/agent-sdk/pull/247
Implement dynamic workspace path detection for str_replace_editor tool by @li-boxuan in https://github.com/All-Hands-AI/agent-sdk/pull/252
add support to customize system prompt via kwargs by @xingyaoww in https://github.com/All-Hands-AI/agent-sdk/pull/257
Bump tj-actions/changed-files from 46 to 47 in the version-all group by @dependabot[bot] in https://github.com/All-Hands-AI/agent-sdk/pull/259
feat: add integration tests and cron job workflow by @simonrosenberg in https://github.com/All-Hands-AI/agent-sdk/pull/219
Consolidate Agent State Management from Multiple Booleans to Single Enum by @xingyaoww in https://github.com/All-Hands-AI/agent-sdk/pull/261
add LLMConvertibleEventWithMetrics and add get_llm_metrics example by @xingyaoww in https://github.com/All-Hands-AI/agent-sdk/pull/262
Support for discriminated unions to dump and re-create from spec by @xingyaoww in https://github.com/All-Hands-AI/agent-sdk/pull/237
port over sdk change related to server implementation by @xingyaoww in https://github.com/All-Hands-AI/agent-sdk/pull/269
Model features: broaden patterns (o1*/o3*/gpt-5*) and small cleanups by @enyst in https://github.com/All-Hands-AI/agent-sdk/pull/273
fix .model_dump_json issue introduced in #237 by @xingyaoww in https://github.com/All-Hands-AI/agent-sdk/pull/277
Revert #262 LLMConvertibleEventWithMetrics to use a new type EventWithMetrics by @xingyaoww in https://github.com/All-Hands-AI/agent-sdk/pull/275
Add LLM JSON storage functionality by @malhotra5 in https://github.com/All-Hands-AI/agent-sdk/pull/278
LLM-summarizing condenser implementation by @csmith49 in https://github.com/All-Hands-AI/agent-sdk/pull/181
Expose methods/dicts for model choices by @malhotra5 in https://github.com/All-Hands-AI/agent-sdk/pull/276
Lift ID types in event to type aliases by @csmith49 in https://github.com/All-Hands-AI/agent-sdk/pull/279
Fix pause logic with new agent status enums by @malhotra5 in https://github.com/All-Hands-AI/agent-sdk/pull/280
Enforce list of Message only for LLM by @ryanhoangt in https://github.com/All-Hands-AI/agent-sdk/pull/260
fix FastAPI serialization with SkipJsonSchema by @xingyaoww in https://github.com/All-Hands-AI/agent-sdk/pull/283
refactor: rename test for consistency by @simonrosenberg in https://github.com/All-Hands-AI/agent-sdk/pull/289
Add t02_add_bash_hello.py integration test by @simonrosenberg in https://github.com/All-Hands-AI/agent-sdk/pull/291
Add t07_interactive_commands.py integration test by @simonrosenberg in https://github.com/All-Hands-AI/agent-sdk/pull/297
Fix integration job when run on schedule by @simonrosenberg in https://github.com/All-Hands-AI/agent-sdk/pull/299
Add t03_jupyter_write_file.py integration test by @simonrosenberg in https://github.com/All-Hands-AI/agent-sdk/pull/292
Add t04_git_staging.py integration test by @simonrosenberg in https://github.com/All-Hands-AI/agent-sdk/pull/293
Replace ConversationID to use uuid.UUID instead of string by @xingyaoww in https://github.com/All-Hands-AI/agent-sdk/pull/286
Integrate browser-use tools by @ryanhoangt in https://github.com/All-Hands-AI/agent-sdk/pull/214
Add t06_github_pr_browsing.py integration test by @simonrosenberg in https://github.com/All-Hands-AI/agent-sdk/pull/295
Add t05_simple_browsing.py integration test by @simonrosenberg in https://github.com/All-Hands-AI/agent-sdk/pull/294
Introduce Spec for simpler "API start" in server by @xingyaoww in https://github.com/All-Hands-AI/agent-sdk/pull/284
stop inherit TextContent from mcp.types.TextContent since it breaks JSON schema by @xingyaoww in https://github.com/All-Hands-AI/agent-sdk/pull/306
Remove CondenserSpec and use Condenser directly in AgentSpec by @xingyaoww in https://github.com/All-Hands-AI/agent-sdk/pull/308
Add LLM costs to integration tests by @simonrosenberg in https://github.com/All-Hands-AI/agent-sdk/pull/304
Ported over agent server code by @tofarr in https://github.com/All-Hands-AI/agent-sdk/pull/317
Add comprehensive documentation and improve configuration for OpenHands Agent Server by @tofarr in https://github.com/All-Hands-AI/agent-sdk/pull/318
Add pre-commit for YAML files by @xingyaoww in https://github.com/All-Hands-AI/agent-sdk/pull/321
Build agent-server into executable binary by @xingyaoww in https://github.com/All-Hands-AI/agent-sdk/pull/320
Add MIT License to the project by @xingyaoww in https://github.com/All-Hands-AI/agent-sdk/pull/325
refactor: remove run_infer.sh for integration tests by @simonrosenberg in https://github.com/All-Hands-AI/agent-sdk/pull/324
Build docker for Agent SDK by @xingyaoww in https://github.com/All-Hands-AI/agent-sdk/pull/326
Comment on PR when docker build finishes by @xingyaoww in https://github.com/All-Hands-AI/agent-sdk/pull/329
Create lazy import statements in tools package by @malhotra5 in https://github.com/All-Hands-AI/agent-sdk/pull/330
refactor: update logs by @simonrosenberg in https://github.com/All-Hands-AI/agent-sdk/pull/332
Bump the version-all group with 3 updates by @dependabot[bot] in https://github.com/All-Hands-AI/agent-sdk/pull/322
Security analyzer framework by @csmith49 in https://github.com/All-Hands-AI/agent-sdk/pull/305
set base_url directly instead of set default for oh provider by @xingyaoww in https://github.com/All-Hands-AI/agent-sdk/pull/338
remove DynamicStrReplaceEditor action by removing the hardcoded path in tool desc by @xingyaoww in https://github.com/All-Hands-AI/agent-sdk/pull/340
Do not add security_risk unless security analyzer is enabled by @xingyaoww in https://github.com/All-Hands-AI/agent-sdk/pull/341
make _add_security_risk_prediction a property for re-use by @xingyaoww in https://github.com/All-Hands-AI/agent-sdk/pull/347
tool: default editor cwd to current cwd by @xingyaoww in https://github.com/All-Hands-AI/agent-sdk/pull/349
Add Docker image variants for Java, Golang by @rbren in https://github.com/All-Hands-AI/agent-sdk/pull/345
Support pass str to send_message in Conversation class by @rbren in https://github.com/All-Hands-AI/agent-sdk/pull/339
Simpler serialization! by @tofarr in https://github.com/All-Hands-AI/agent-sdk/pull/331
Fix CI: Only run build-and-push-image on main or build-docker tag by @xingyaoww in https://github.com/All-Hands-AI/agent-sdk/pull/361
Create MCPToolAction with dynamic schema by @xingyaoww in https://github.com/All-Hands-AI/agent-sdk/pull/342
add a simple conversation viewer for debugging purpose by @xingyaoww in https://github.com/All-Hands-AI/agent-sdk/pull/363
Add gte constraints to Pydantic numeric fields by @tofarr in https://github.com/All-Hands-AI/agent-sdk/pull/367
Fix for serialization by @tofarr in https://github.com/All-Hands-AI/agent-sdk/pull/364
Use Blacksmith runners for all GitHub Actions workflows by @xingyaoww in https://github.com/All-Hands-AI/agent-sdk/pull/370
Bump the version-all group across 1 directory with 5 updates by @dependabot[bot] in https://github.com/All-Hands-AI/agent-sdk/pull/359
feat: add expose_secrets context parameter for AgentSpec serialization by @malhotra5 in https://github.com/All-Hands-AI/agent-sdk/pull/368
More serialization fix by @xingyaoww in https://github.com/All-Hands-AI/agent-sdk/pull/374
docs(spec): fix AgentSpec examples in comments and align README with examples by @enyst in https://github.com/All-Hands-AI/agent-sdk/pull/380
docs: sync README with examples and presets by @enyst in https://github.com/All-Hands-AI/agent-sdk/pull/381
Relax openapi schema test & make Microagent a DiscriminatedUnionMixin and tests by @xingyaoww in https://github.com/All-Hands-AI/agent-sdk/pull/377
Unify tool initialization — Step 1: global registry + lazy materialization by @xingyaoww in https://github.com/All-Hands-AI/agent-sdk/pull/356
Update README.md to match current codebase structure and API by @xingyaoww in https://github.com/All-Hands-AI/agent-sdk/pull/384
docs(repo): add section 'Avoid overly defensive code' to repo.md by @enyst in https://github.com/All-Hands-AI/agent-sdk/pull/385
chore(typing): adopt Python 3.12 style (PEP 585/604) by @enyst in https://github.com/All-Hands-AI/agent-sdk/pull/382
refactor: centralize tool conversion in LLM (split from #142) by @enyst in https://github.com/All-Hands-AI/agent-sdk/pull/371
chore(visualization): remove dim everywhere, fix diff newline by @xingyaoww in https://github.com/All-Hands-AI/agent-sdk/pull/389
Add example showing agent can handle messages sent while task is running by @simonrosenberg in https://github.com/All-Hands-AI/agent-sdk/pull/378
Move tests importing openhands.tools from tests/sdk to tests/cross by @xingyaoww in https://github.com/All-Hands-AI/agent-sdk/pull/390
Replaces confirmation mode with confirmation policy by @csmith49 in https://github.com/All-Hands-AI/agent-sdk/pull/348
server: register default tools on server & add /tool/list endpoint for registry by @xingyaoww in https://github.com/All-Hands-AI/agent-sdk/pull/392
Server details by @tofarr in https://github.com/All-Hands-AI/agent-sdk/pull/362
Fix server hang due to browser-use & more loggings for server by @xingyaoww in https://github.com/All-Hands-AI/agent-sdk/pull/399
Add LLM router by @ryanhoangt in https://github.com/All-Hands-AI/agent-sdk/pull/152
Disable auth for /server_info endpoint by @rbren in https://github.com/All-Hands-AI/agent-sdk/pull/405
Fix: Reverse --reload flag logic to default to False by @rbren in https://github.com/All-Hands-AI/agent-sdk/pull/406
Add sudo to base package installation in Dockerfile by @rbren in https://github.com/All-Hands-AI/agent-sdk/pull/409
Add optional static files directory support to agent server by @tofarr in https://github.com/All-Hands-AI/agent-sdk/pull/395
Change default base image to slim version by @rbren in https://github.com/All-Hands-AI/agent-sdk/pull/404
Add VSCode and Docker to agent-server Dockerfile by @rbren in https://github.com/All-Hands-AI/agent-sdk/pull/412
Build Docker images for Python on all commits by @rbren in https://github.com/All-Hands-AI/agent-sdk/pull/413
Remove custom model_dump_with_secrets method and use Pydantic context-based serialization by @malhotra5 in https://github.com/All-Hands-AI/agent-sdk/pull/423
chore: remove --no-reload that was removed by @xingyaoww in https://github.com/All-Hands-AI/agent-sdk/pull/424
chore(log): flip logger.info to logger.debug for LLM by @xingyaoww in https://github.com/All-Hands-AI/agent-sdk/pull/425
Consolidate security content into centralized security policy by @neubig in https://github.com/All-Hands-AI/agent-sdk/pull/422
Port over stuck detector by @ryanhoangt in https://github.com/All-Hands-AI/agent-sdk/pull/335
Prefix api routes with /api by @tofarr in https://github.com/All-Hands-AI/agent-sdk/pull/428
Fix persistence path in default agent by @malhotra5 in https://github.com/All-Hands-AI/agent-sdk/pull/429
chore(log): make condenser log debug by @xingyaoww in https://github.com/All-Hands-AI/agent-sdk/pull/430
Handle context window exceeded exceptions with registered condenser by @csmith49 in https://github.com/All-Hands-AI/agent-sdk/pull/358
fix condensation issue caused by condenser view by @xingyaoww in https://github.com/All-Hands-AI/agent-sdk/pull/432
Expand webhook protocol to support conversation lifecycle events by @tofarr in https://github.com/All-Hands-AI/agent-sdk/pull/379
Bump actions/setup-node from 4 to 5 in the version-all group by @dependabot[bot] in https://github.com/All-Hands-AI/agent-sdk/pull/401
refactor: create ObservationBaseEvent and refactor llm_convertible events to multiple files by @xingyaoww in https://github.com/All-Hands-AI/agent-sdk/pull/433
chore: simplify examples/04_confirmation_mode_example.py by @xingyaoww in https://github.com/All-Hands-AI/agent-sdk/pull/436
Fix security analyzer by @xingyaoww in https://github.com/All-Hands-AI/agent-sdk/pull/435
refactor(conversation): initial protocol for conversation by @xingyaoww in https://github.com/All-Hands-AI/agent-sdk/pull/431
Fix exception handling for unauthorized requests by @rbren in https://github.com/All-Hands-AI/agent-sdk/pull/408
Documentation updates by @tofarr in https://github.com/All-Hands-AI/agent-sdk/pull/434
Fix websocket infinite loop on user disconnect by @tofarr in https://github.com/All-Hands-AI/agent-sdk/pull/439
Solve final action concurrent message bug by @simonrosenberg in https://github.com/All-Hands-AI/agent-sdk/pull/403
Authentication Upgrades by @tofarr in https://github.com/All-Hands-AI/agent-sdk/pull/438
Add stale issue and PR management workflow by @xingyaoww in https://github.com/All-Hands-AI/agent-sdk/pull/445
Added file upload download routes and reorganized endpoints using tags by @tofarr in https://github.com/All-Hands-AI/agent-sdk/pull/448
feat(sdk): Implement LocalConversation and RemoteConversation (fixes #388) by @xingyaoww in https://github.com/All-Hands-AI/agent-sdk/pull/391
Port over conversation stats by @malhotra5 in https://github.com/All-Hands-AI/agent-sdk/pull/288
Expose property for when confirmation policy is set by @malhotra5 in https://github.com/All-Hands-AI/agent-sdk/pull/449
Add LLMSecurityAnalyzer as default security analyzer in get_default_agent by @malhotra5 in https://github.com/All-Hands-AI/agent-sdk/pull/454
Docker Sandboxed Remote Conversation by @xingyaoww in https://github.com/All-Hands-AI/agent-sdk/pull/455
Add Example Application by @tofarr in https://github.com/All-Hands-AI/agent-sdk/pull/393
Fix concurrent user messages sent while agent is finishing by @simonrosenberg in https://github.com/All-Hands-AI/agent-sdk/pull/443
chore(ruff): enforce Python 3.12 typing standards with UP006, UP007, … by @simonrosenberg in https://github.com/All-Hands-AI/agent-sdk/pull/464
Removed conversation_id from event api by @tofarr in https://github.com/All-Hands-AI/agent-sdk/pull/457
Removed reference to old name by @tofarr in https://github.com/All-Hands-AI/agent-sdk/pull/466
Another rename by @tofarr in https://github.com/All-Hands-AI/agent-sdk/pull/467
Add pyupgrade rules to Ruff config by @xingyaoww in https://github.com/All-Hands-AI/agent-sdk/pull/465
fix sonnet 4 output size by @xingyaoww in https://github.com/All-Hands-AI/agent-sdk/pull/468
Add Chromium binary check and auto-install if not available by @ryanhoangt in https://github.com/All-Hands-AI/agent-sdk/pull/440
Feat: add cron job for typing quality check by @simonrosenberg in https://github.com/All-Hands-AI/agent-sdk/pull/462
Fix typing issues: remove vague unions, optional runtime fields, and consumer-side narrowing by @simonrosenberg in https://github.com/All-Hands-AI/agent-sdk/pull/461
Add comprehensive tests for browser-use tools by @ryanhoangt in https://github.com/All-Hands-AI/agent-sdk/pull/441
Fix: makes sure conversation stats is passed to conversation visualizer by @malhotra5 in https://github.com/All-Hands-AI/agent-sdk/pull/478
Tighten typing to remove vague unions and optional runtime fields by @simonrosenberg in https://github.com/All-Hands-AI/agent-sdk/pull/473
Fix #470: Update PR description instead of creating comments for agent server images by @xingyaoww in https://github.com/All-Hands-AI/agent-sdk/pull/472
Improve server logging for HTTP 500 by @xingyaoww in https://github.com/All-Hands-AI/agent-sdk/pull/480
feat: Add BashTaskService for executing bash commands outside agent context by @tofarr in https://github.com/All-Hands-AI/agent-sdk/pull/453
Refactor LLM registry to remove service_id parameter from add() method by @malhotra5 in https://github.com/All-Hands-AI/agent-sdk/pull/482
Move WebSocket endpoints to /sockets and use query parameter authentication by @tofarr in https://github.com/All-Hands-AI/agent-sdk/pull/486
Fix quoted 'Self' type annotations by @simonrosenberg in https://github.com/All-Hands-AI/agent-sdk/pull/488
Add reset functionality to BashExecutor for unresponsive terminals by @xingyaoww in https://github.com/All-Hands-AI/agent-sdk/pull/474
Fix server.yml PR description to account for all Docker variants by @xingyaoww in https://github.com/All-Hands-AI/agent-sdk/pull/483
Add @abstractmethod decorators to base classes by @simonrosenberg in https://github.com/All-Hands-AI/agent-sdk/pull/490
Remove pandas from dev dependencies by @xingyaoww in https://github.com/All-Hands-AI/agent-sdk/pull/496
Public APIs return OpenHands types not LiteLLM types - Part 1: CompletionResult by @jpshackelford in https://github.com/All-Hands-AI/agent-sdk/pull/375
feat: Add Web VSCode support to AgentServer by @xingyaoww in https://github.com/All-Hands-AI/agent-sdk/pull/477
Fix LLM service_id naming inconsistencies in examples by @malhotra5 in https://github.com/All-Hands-AI/agent-sdk/pull/499
Update dockerfile with browser-use deps by @ryanhoangt in https://github.com/All-Hands-AI/agent-sdk/pull/502
fix remote conversation by update socket url; add browsing example by @xingyaoww in https://github.com/All-Hands-AI/agent-sdk/pull/505
Serve server info from / rather than 404 if no static files by @tofarr in https://github.com/All-Hands-AI/agent-sdk/pull/512
Fix sandbox build script lookup to work from any directory by @juanmichelini in https://github.com/All-Hands-AI/agent-sdk/pull/513
Fix PyInstaller binary build: include all Jinja2 template files by @xingyaoww in https://github.com/All-Hands-AI/agent-sdk/pull/514
Add API key support to RemoteConversation by @rbren in https://github.com/All-Hands-AI/agent-sdk/pull/523
Remove load_from_toml function and tomli dependency by @xingyaoww in https://github.com/All-Hands-AI/agent-sdk/pull/526
Fix sonnet3.5 recognition by @enyst in https://github.com/All-Hands-AI/agent-sdk/pull/524
feat: Add generic environment variable configuration parsing for Pydantic models by @tofarr in https://github.com/All-Hands-AI/agent-sdk/pull/511
Include conversation_id parameter in WebhookSubscriber by @tofarr in https://github.com/All-Hands-AI/agent-sdk/pull/528
Refactor: Move presets to tools package and integrate LLM metadata by @xingyaoww in https://github.com/All-Hands-AI/agent-sdk/pull/532
Refactor: Fix agent to continue conversation when tool doesn't exist by @simonrosenberg in https://github.com/All-Hands-AI/agent-sdk/pull/546
Refactor: move get_unmatched_actions to ConversationState as static method by @simonrosenberg in https://github.com/All-Hands-AI/agent-sdk/pull/540
refactor: remove llm context manager by @simonrosenberg in https://github.com/All-Hands-AI/agent-sdk/pull/537
Support VNC inside agent-server - and allows for interactive browsing!! by @xingyaoww in https://github.com/All-Hands-AI/agent-sdk/pull/515
Refactor: remove pass statements from abstract methods and add meaningful docstrings by @simonrosenberg in https://github.com/All-Hands-AI/agent-sdk/pull/542
add claude-sonnet-4.5 by @xingyaoww in https://github.com/All-Hands-AI/agent-sdk/pull/558
Fix remaining claude-4-5 issues by @xingyaoww in https://github.com/All-Hands-AI/agent-sdk/pull/560
Bump the version-all group with 3 updates by @dependabot[bot] in https://github.com/All-Hands-AI/agent-sdk/pull/536
Support Two Types of Directories for Conversations: persistence_dir and working_dir by @xingyaoww in https://github.com/All-Hands-AI/agent-sdk/pull/531
Refactor: deduplicate compose_callbacks by @simonrosenberg in https://github.com/All-Hands-AI/agent-sdk/pull/568
Replace ListLike Protocol with list[T] type annotations by @simonrosenberg in https://github.com/All-Hands-AI/agent-sdk/pull/562
Implement Workspace and support system-level bash execution by @xingyaoww in https://github.com/All-Hands-AI/agent-sdk/pull/534
Fix FIFOLock + pytest collection warnings by renaming test helper classes by @simonrosenberg in https://github.com/All-Hands-AI/agent-sdk/pull/572
Add ConversationState updates to websocket events by @xingyaoww in https://github.com/All-Hands-AI/agent-sdk/pull/556
fix(sdk): prevent CLI crash by early-validating MCP tool args (closes All-Hands-AI/OpenHands#11163) by @neubig in https://github.com/All-Hands-AI/agent-sdk/pull/529
fix: changed suffix in agent server naming from timestamp from random uuid by @juanmichelini in https://github.com/All-Hands-AI/agent-sdk/pull/577
Fix FIFOLock test race condition and pytest warnings by @simonrosenberg in https://github.com/All-Hands-AI/agent-sdk/pull/575
Have post to /conversations work without redirect by @tofarr in https://github.com/All-Hands-AI/agent-sdk/pull/578
Creating a workspace object should not try to create the dir by @tofarr in https://github.com/All-Hands-AI/agent-sdk/pull/579
auto-create working dir if it doesn't exists for LocalConversation by @xingyaoww in https://github.com/All-Hands-AI/agent-sdk/pull/582
add py.typed for all modules by @xingyaoww in https://github.com/All-Hands-AI/agent-sdk/pull/584
Pillow now seems to be required in the agent server by @tofarr in https://github.com/All-Hands-AI/agent-sdk/pull/580
add Agent to ConversationStateProtocol by @xingyaoww in https://github.com/All-Hands-AI/agent-sdk/pull/586
DockerFile. now detects correct chromium name for Ubuntu and Debian by @juanmichelini in https://github.com/All-Hands-AI/agent-sdk/pull/588
Add Anthropic thinking blocks support to events system by @xingyaoww in https://github.com/All-Hands-AI/agent-sdk/pull/576
feat: env path to docker building files not in packages by @juanmichelini in https://github.com/All-Hands-AI/agent-sdk/pull/587
Fix multiple inheritance issue in ConversationState (#583) by @xingyaoww in https://github.com/All-Hands-AI/agent-sdk/pull/585
Fix nested LLM reconciliation in agent deserialization by @malhotra5 in https://github.com/All-Hands-AI/agent-sdk/pull/517
Enable interleaved thinking by @xingyaoww in https://github.com/All-Hands-AI/agent-sdk/pull/594
Fix sonnet 4.5 output token by @xingyaoww in https://github.com/All-Hands-AI/agent-sdk/pull/595
Refactor OpenHands SDK: Optional service_id, rename base classes, and enhance Conversation by @rbren in https://github.com/All-Hands-AI/agent-sdk/pull/591
Rename Tool class to ToolDefinition across entire codebase by @rbren in https://github.com/All-Hands-AI/agent-sdk/pull/592
Remove Tool = ToolDefinition alias and Tool exports by @xingyaoww in https://github.com/All-Hands-AI/agent-sdk/pull/602
Add ZIP download for conversation visualizer and fix directory reset issue by @xingyaoww in https://github.com/All-Hands-AI/agent-sdk/pull/601
Rename ToolSpec to Tool throughout the repository by @rbren in https://github.com/All-Hands-AI/agent-sdk/pull/599
Remove deprecated params={'working_dir'} and params={'workspace_root'} patterns from tools by @xingyaoww in https://github.com/All-Hands-AI/agent-sdk/pull/604
Fix MCP tool security risk prediction schema bug by @xingyaoww in https://github.com/All-Hands-AI/agent-sdk/pull/607
Fix StrReplaceEditor to include working directory in tool description by @simonrosenberg in https://github.com/All-Hands-AI/agent-sdk/pull/605
Add tests for Conversation constructor with secrets parameter by @rbren in https://github.com/All-Hands-AI/agent-sdk/pull/597
Better implementation of get_event by @tofarr in https://github.com/All-Hands-AI/agent-sdk/pull/610
Add resend_all parameter to events_socket endpoint by @tofarr in https://github.com/All-Hands-AI/agent-sdk/pull/608
Fix thinking budget 200k raising error with interleaved thinking by @ryanhoangt in https://github.com/All-Hands-AI/agent-sdk/pull/614
Add Pydantic models for workspace method return values by @tofarr in https://github.com/All-Hands-AI/agent-sdk/pull/615
feat: Add DockerWorkspace by @xingyaoww in https://github.com/All-Hands-AI/agent-sdk/pull/510
fix event service duplicate id by @xingyaoww in https://github.com/All-Hands-AI/agent-sdk/pull/617
Formal secrets by @tofarr in https://github.com/All-Hands-AI/agent-sdk/pull/616
Refactor: add MessageToolCall class instead of litellm's ChatCompletionMessageToolCall by @simonrosenberg in https://github.com/All-Hands-AI/agent-sdk/pull/550
Add full agent/LLM logs to integration test artifacts by @simonrosenberg in https://github.com/All-Hands-AI/agent-sdk/pull/612
Can now specify secrets on conversation start, and simplified type hints by @tofarr in https://github.com/All-Hands-AI/agent-sdk/pull/625
Refactor: rename str_replace_editor -> file_editor by @simonrosenberg in https://github.com/All-Hands-AI/agent-sdk/pull/628
Allow FileEditor insert_line to be 0 (start-of-file) by @enyst in https://github.com/All-Hands-AI/agent-sdk/pull/635
feat: NonExecutableActionEvent for Completions path (preserve tool_call_id continuity) by @enyst in https://github.com/All-Hands-AI/agent-sdk/pull/623
Add native responses by @enyst in https://github.com/All-Hands-AI/agent-sdk/pull/622
LLM → Telemetry: fixes by @enyst in https://github.com/All-Hands-AI/agent-sdk/pull/638
Remove deprecated to_llm_dict() and from_litellm_message() methods by @xingyaoww in https://github.com/All-Hands-AI/agent-sdk/pull/642
Bump the version-all group with 2 updates by @dependabot[bot] in https://github.com/All-Hands-AI/agent-sdk/pull/645
Fix: Remove 'kind' field from MCP schema output by @xingyaoww in https://github.com/All-Hands-AI/agent-sdk/pull/641
Implement generate_title method for Conversation class by @xingyaoww in https://github.com/All-Hands-AI/agent-sdk/pull/256
Revise README for OpenHands Agent SDK by @neubig in https://github.com/All-Hands-AI/agent-sdk/pull/649
Encapsulate conversation title generation via EventService by @enyst in https://github.com/All-Hands-AI/agent-sdk/pull/653
Remove automatic Chromium installation from browser tool by @rbren in https://github.com/All-Hands-AI/agent-sdk/pull/652
Replace LITELLM_API_KEY with LLM_API_KEY in examples by @neubig in https://github.com/All-Hands-AI/agent-sdk/pull/654
Support GPT-5-Codex by @xingyaoww in https://github.com/All-Hands-AI/agent-sdk/pull/655
Fix browser use tools not init error by @ryanhoangt in https://github.com/All-Hands-AI/agent-sdk/pull/658
refactor: move DockerWorkspace to separate openhands/workspace package by @rbren in https://github.com/All-Hands-AI/agent-sdk/pull/666
Feat: make read-only planning agent by @simonrosenberg in https://github.com/All-Hands-AI/agent-sdk/pull/624
Investigate Integration Tests Failures by @simonrosenberg in https://github.com/All-Hands-AI/agent-sdk/pull/651
fix: prevent tests from making HTTP calls to production LLM proxy by @neubig in https://github.com/All-Hands-AI/agent-sdk/pull/673
Refactor: integration tests by @simonrosenberg in https://github.com/All-Hands-AI/agent-sdk/pull/677
fix(backend): [agent server] remote workspace loading all bash events instead of using filters by @hieptl in https://github.com/All-Hands-AI/agent-sdk/pull/679
Allow security_analyzer to differ during agent reconciliation by @xingyaoww in https://github.com/All-Hands-AI/agent-sdk/pull/669
fix: Path(file) points to workspace.py not docker.py by @juanmichelini in https://github.com/All-Hands-AI/agent-sdk/pull/683
Add is_confirmation_mode_active property to BaseConversation by @malhotra5 in https://github.com/All-Hands-AI/agent-sdk/pull/687
Fix for small logic hole in env parser by @tofarr in https://github.com/All-Hands-AI/agent-sdk/pull/688
Remote Runtime API (no build) by @xingyaoww in https://github.com/All-Hands-AI/agent-sdk/pull/684
feat(backend): [Agent Server] Implement update_conversation API by @hieptl in https://github.com/All-Hands-AI/agent-sdk/pull/661
Refactor Docker build system: Replace variant with custom_tags and prioritize SHA-based tags by @xingyaoww in https://github.com/All-Hands-AI/agent-sdk/pull/685
Support building "minimal" version of agent-server docker image by @xingyaoww in https://github.com/All-Hands-AI/agent-sdk/pull/693
Bug fix gpt 5 responses api bis by @simonrosenberg in https://github.com/All-Hands-AI/agent-sdk/pull/699
fix color in makefile by @xingyaoww in https://github.com/All-Hands-AI/agent-sdk/pull/705
fix coverage report comment by making it shorter by @raymyers in https://github.com/All-Hands-AI/agent-sdk/pull/701
Reorganize examples by @xingyaoww in https://github.com/All-Hands-AI/agent-sdk/pull/696
remote convo: log request error by @xingyaoww in https://github.com/All-Hands-AI/agent-sdk/pull/709
Python-based Docker Build Script by @xingyaoww in https://github.com/All-Hands-AI/agent-sdk/pull/698
Fix ConversationService lifecycle management to prevent inactive_service errors by @tofarr in https://github.com/All-Hands-AI/agent-sdk/pull/715
Add configurable VSCode port parameter by @tofarr in https://github.com/All-Hands-AI/agent-sdk/pull/717
Bump astral-sh/setup-uv from 6 to 7 in the version-all group by @dependabot[bot] in https://github.com/All-Hands-AI/agent-sdk/pull/721
Update planning system prompt by @simonrosenberg in https://github.com/All-Hands-AI/agent-sdk/pull/719
Update scripts to fix websocket by @tofarr in https://github.com/All-Hands-AI/agent-sdk/pull/716
Remove AGENT_SDK_PATH environment variable by @juanmichelini in https://github.com/All-Hands-AI/agent-sdk/pull/724
Add Vscode initial settings extension by @enyst in https://github.com/All-Hands-AI/agent-sdk/pull/720
Runtime errors with security risk and no configured security analyzer by @malhotra5 in https://github.com/All-Hands-AI/agent-sdk/pull/723
Refactor hello_world.py for simplicity by @neubig in https://github.com/All-Hands-AI/agent-sdk/pull/713
use older version of builder so the binary is compatible by @xingyaoww in https://github.com/All-Hands-AI/agent-sdk/pull/731
Added run flag to send message operations by @tofarr in https://github.com/All-Hands-AI/agent-sdk/pull/729
Fix agent server persistence by @tofarr in https://github.com/All-Hands-AI/agent-sdk/pull/732
Add gpt-5-codex to OpenHands verified models by @xingyaoww in https://github.com/All-Hands-AI/agent-sdk/pull/657
v1.0.0a1 Release by @xingyaoww in https://github.com/All-Hands-AI/agent-sdk/pull/736

New Contributors

@malhotra5 made their first contribution in https://github.com/All-Hands-AI/agent-sdk/pull/69
@ryanhoangt made their first contribution in https://github.com/All-Hands-AI/agent-sdk/pull/70
@csmith49 made their first contribution in https://github.com/All-Hands-AI/agent-sdk/pull/61
@dependabot[bot] made their first contribution in https://github.com/All-Hands-AI/agent-sdk/pull/100
@simonrosenberg made their first contribution in https://github.com/All-Hands-AI/agent-sdk/pull/178
@tofarr made their first contribution in https://github.com/All-Hands-AI/agent-sdk/pull/217
@li-boxuan made their first contribution in https://github.com/All-Hands-AI/agent-sdk/pull/252
@rbren made their first contribution in https://github.com/All-Hands-AI/agent-sdk/pull/345
@neubig made their first contribution in https://github.com/All-Hands-AI/agent-sdk/pull/422
@jpshackelford made their first contribution in https://github.com/All-Hands-AI/agent-sdk/pull/375
@juanmichelini made their first contribution in https://github.com/All-Hands-AI/agent-sdk/pull/513
@hieptl made their first contribution in https://github.com/All-Hands-AI/agent-sdk/pull/679
@raymyers made their first contribution in https://github.com/All-Hands-AI/agent-sdk/pull/701

Full Changelog: https://github.com/All-Hands-AI/agent-sdk/commits/1.0.0a1

Open source →

Your coding agent can read these notes before it upgrades. Set up the MCP server →