NewYour coding agent can read the release notes before it upgrades.Set up the MCP server →
Go modules, #697 by repository stars
Latest stable
v0.2.0
Last release 1 months ago
14 Oct 2026
Ships on a steady schedule
a new release about every 9 days
Nearly every release is documented
notes for 57 of the last 60 stable releases
Nothing withdrawn
no release was ever pulled
1 years old
929 releases · first in 2025
Drag across the ruler to compare, or tap a release to open it.
security(deps): bump pytest to >=9.0.3 (CVE-2025-71176) by @AbirAbbas in https://github.com/Agent-Field/agentfield/pull/488
⚠️ This is a staging/pre-release version for testing. Not recommended for production use.
# Staging binary (use --staging flag)
curl -fsSL https://agentfield.ai/install.sh | bash -s -- --staging
# Python SDK (prerelease - requires --pre flag)
pip install --pre agentfield
# TypeScript SDK
npm install @agentfield/sdk@next
VERSION=v0.1.140-rc.2 curl -fsSL https://agentfield.ai/install.sh | bash
Download the binary for your platform below, make it executable, and move it to your PATH.
agentfield-darwin-amd64agentfield-darwin-arm64agentfield-linux-amd64agentfield-linux-arm64af call rejecting valid input for optional reasoner params by @AbirAbbas in https://github.com/Agent-Field/agentfield/pull/610_current_status issue where status stuck on `S… by @DebanKsahu in https://github.com/Agent-Field/agentfield/pull/673af install by @AbirAbbas in https://github.com/Agent-Field/agentfield/pull/738Note truncated.
Your coding agent can read these notes before it upgrades. Set up the MCP server →
The Go OpenCode provider folded the harness system prompt into the positional
opencode run prompt and passed no agent, so a run depended on whatever agent
and permissions the ambient OpenCode config happened to define. Python moved off
that in #1023; this brings Go to the same behaviour.
Each run now selects the fixed agentfield-harness agent with --agent and
supplies it through OPENCODE_CONFIG_CONTENT: system prompt, model,
reasoningEffort, mode primary, a fixed steps budget, and a headless permission
baseline that denies question, task and the agentfield* skills so an
AgentField-launched worker cannot dispatch back into the control plane.
The overlay is deep-merged into the caller's per-call value, or the ambient one when there is no per-call value, so a deployment's mcp servers, plugins, providers and other agents survive and the OpenRouter attribution overlay and the harness agent coexist. A caller value that is not a JSON object fails the run before the concurrency slot is taken and before the child is launched.
AGENTFIELD_OPENCODE_INLINE_SYSTEM_PROMPT restores the inline prompt transport
and strips the agent's configured prompt, keeping the agent selection and
permissions, so a caller with a very long system prompt can roll back without
pinning an older SDK. tools and permission_mode stay untranslated: with the
wildcard allow in place a tool mapping would only write allow on top of allow.
steps is a named constant with an AGENTFIELD_OPENCODE_STEPS override and is
never fed from max_turns.
Refs #960
Co-Authored-By: Claude Opus 5 (1M context) noreply@anthropic.com
Mirrors the Go and Python OpenCode providers: each run selects the fixed
agentfield-harness agent with --agent and defines it through a per-run
OPENCODE_CONFIG_CONTENT overlay (system prompt, model, reasoningEffort, mode
primary, fixed steps, and the headless permission baseline that denies
question, task and the agentfield* skills). The task is the only
positional prompt.
The overlay is deep-merged into the caller's per-call value or the ambient one, so deployment-owned mcp servers, plugins and agents survive and the OpenRouter attribution overlay is no longer the only thing that can occupy the variable. Object key order is preserved deliberately, with the wildcard first and AgentField's denials last, because OpenCode applies the last matching rule. A caller value that is not a JSON object throws before runCli is called.
AGENTFIELD_OPENCODE_INLINE_SYSTEM_PROMPT restores the inline prompt transport,
and tools / permission_mode remain accepted but untranslated.
This also fixes a key-name bug the overlay would otherwise inherit: the
provider read options.system_prompt, but HarnessRunner forwards HarnessOptions
verbatim, so a system prompt set through the public TypeScript API arrived as
systemPrompt and never reached opencode at all. It now accepts both spellings,
as the aforge provider already does.
Refs #960
Co-Authored-By: Claude Opus 5 (1M context) noreply@anthropic.com
The OpenCode agent overlay hard-coded steps: 500 as a bare literal. Give it a
name and an AGENTFIELD_OPENCODE_STEPS override (per-call environment first, then
ambient; non-numeric, zero and negative values fall back to the default), so all
three SDKs expose the same knob. The default is unchanged and max_turns is
still never serialized as steps.
Refs #960
Co-Authored-By: Claude Opus 5 (1M context) noreply@anthropic.com
The standalone-runs section was written when only Python had the per-run agent overlay, and the provider-parity table still said OpenCode receives only the model, directory and prompt. Both are now true of Go and TypeScript too.
Also documents AGENTFIELD_OPENCODE_STEPS and states plainly that tools and
permission_mode are accepted and ignored, rather than leaving readers to infer
they are wired to OpenCode permissions.
Refs #960
Co-Authored-By: Claude Opus 5 (1M context) noreply@anthropic.com
OpenCode applies the last matching permission rule, so the Python and
TypeScript providers deliberately emit AgentField's skill/question/task denials
after any caller rules. Go relied on encoding/json sorting map keys, which is
only accidentally correct: a caller supplying
{"agent":{"agentfield-harness":{"permission":{"skill":{"agentfield-*":"allow"}}}}}
serialized as {"agentfield*":"deny","agentfield-*":"allow"} — * sorts before
- — so the caller's allow was the last match and the recursion guard was off.
Serialize the permission object through a small ordered JSON type instead, in
the same order Python uses: wildcard, caller rules, then AgentField's denials,
with agentfield* last inside skill. Both the merged and the generated-only
paths now go through it, so one mechanism governs the order.
Also sizes the deep-merge map from the base alone; summing both lengths is what CodeQL's allocation-size-overflow rule flags.
The two new tests assert on the serialized JSON rather than a decoded map, because a decoded map cannot express order; both fail against the previous implementation.
Refs #960
Co-Authored-By: Claude Opus 5 (1M context) noreply@anthropic.com
HarnessRunner forwards HarnessOptions verbatim, so a caller sets
permissionMode and maxTurns; the test only passed the snake_case aliases, so
"tools and permission_mode add nothing to the overlay" was not actually checked
against the keys the public API sends. Pass both spellings.
Also scopes the Windows stdin sentence in the harness docs: Python and Go send the prompt over stdin there, the TypeScript adapter always uses the positional argument. That difference is pre-existing and stays out of this change.
Refs #960
Co-Authored-By: Claude Opus 5 (1M context) noreply@anthropic.com
Co-authored-by: Claude Opus 5 (1M context) noreply@anthropic.com (f7bae4b)
security(deps): bump pytest to >=9.0.3 (CVE-2025-71176) by @AbirAbbas in https://github.com/Agent-Field/agentfield/pull/488
⚠️ This is a staging/pre-release version for testing. Not recommended for production use.
# Staging binary (use --staging flag)
curl -fsSL https://agentfield.ai/install.sh | bash -s -- --staging
# Python SDK (prerelease - requires --pre flag)
pip install --pre agentfield
# TypeScript SDK
npm install @agentfield/sdk@next
VERSION=v0.1.140-rc.1 curl -fsSL https://agentfield.ai/install.sh | bash
Download the binary for your platform below, make it executable, and move it to your PATH.
agentfield-darwin-amd64agentfield-darwin-arm64agentfield-linux-amd64agentfield-linux-arm64af call rejecting valid input for optional reasoner params by @AbirAbbas in https://github.com/Agent-Field/agentfield/pull/610_current_status issue where status stuck on `S… by @DebanKsahu in https://github.com/Agent-Field/agentfield/pull/673af install by @AbirAbbas in https://github.com/Agent-Field/agentfield/pull/738Note truncated.
Bumps the uv group with 1 update in the /sdk/python directory: anyio.
Updates anyio from 4.13.0 to 4.14.2
updated-dependencies:
Signed-off-by: dependabot[bot] support@github.com Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com> (4fc037f)
issue/reaper-retry-child-guard: mirror reaper parent/child and approval guards in stale retry
chore(reaper-retry-child-guard): checkpoint uncommitted issue work (c5407ca)
fix(security): close open Dependabot vulnerability alerts by @santoshkumarradha in https://github.com/Agent-Field/agentfield/pull/1055
curl -fsSL https://agentfield.ai/install.sh | bash
VERSION=v0.1.139 curl -fsSL https://agentfield.ai/install.sh | bash
Download the binary for your platform below, make it executable, and move it to your PATH.
agentfield-darwin-amd64agentfield-darwin-arm64agentfield-linux-amd64agentfield-linux-arm64Full Changelog: https://github.com/Agent-Field/agentfield/compare/v0.1.138...v0.1.139
security(deps): bump pytest to >=9.0.3 (CVE-2025-71176) by @AbirAbbas in https://github.com/Agent-Field/agentfield/pull/488
⚠️ This is a staging/pre-release version for testing. Not recommended for production use.
# Staging binary (use --staging flag)
curl -fsSL https://agentfield.ai/install.sh | bash -s -- --staging
# Python SDK (prerelease - requires --pre flag)
pip install --pre agentfield
# TypeScript SDK
npm install @agentfield/sdk@next
VERSION=v0.1.139-rc.6 curl -fsSL https://agentfield.ai/install.sh | bash
Download the binary for your platform below, make it executable, and move it to your PATH.
agentfield-darwin-amd64agentfield-darwin-arm64agentfield-linux-amd64agentfield-linux-arm64af call rejecting valid input for optional reasoner params by @AbirAbbas in https://github.com/Agent-Field/agentfield/pull/610_current_status issue where status stuck on `S… by @DebanKsahu in https://github.com/Agent-Field/agentfield/pull/673af install by @AbirAbbas in https://github.com/Agent-Field/agentfield/pull/738Note truncated.
Bumps the go_modules group with 2 updates in the /control-plane directory: go.opentelemetry.io/otel/sdk and google.golang.org/grpc.
Updates go.opentelemetry.io/otel/sdk from 1.44.0 to 1.46.0
Updates google.golang.org/grpc from 1.83.1 to 1.83.2
updated-dependencies:
Signed-off-by: dependabot[bot] support@github.com
OpenTelemetry 1.46 stopped appending the default trace signal path for WithEndpointURL (open-telemetry/opentelemetry-go#8538). Normalize pathless and root full HTTP endpoints to /v1/traces before constructing the exporter while preserving explicit collector paths and leaving gRPC behavior unchanged.
Cover the root, trailing-slash, and custom-path cases so future dependency updates cannot silently change the request target again.
Co-Authored-By: Claude Fable 5.1 noreply@anthropic.com
Signed-off-by: dependabot[bot] support@github.com Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com> Co-authored-by: Abir Abbas abirabbas1998@gmail.com Co-authored-by: Claude Fable 5.1 noreply@anthropic.com (c7bcce9)
security(deps): bump pytest to >=9.0.3 (CVE-2025-71176) by @AbirAbbas in https://github.com/Agent-Field/agentfield/pull/488
⚠️ This is a staging/pre-release version for testing. Not recommended for production use.
# Staging binary (use --staging flag)
curl -fsSL https://agentfield.ai/install.sh | bash -s -- --staging
# Python SDK (prerelease - requires --pre flag)
pip install --pre agentfield
# TypeScript SDK
npm install @agentfield/sdk@next
VERSION=v0.1.139-rc.5 curl -fsSL https://agentfield.ai/install.sh | bash
Download the binary for your platform below, make it executable, and move it to your PATH.
agentfield-darwin-amd64agentfield-darwin-arm64agentfield-linux-amd64agentfield-linux-arm64af call rejecting valid input for optional reasoner params by @AbirAbbas in https://github.com/Agent-Field/agentfield/pull/610_current_status issue where status stuck on `S… by @DebanKsahu in https://github.com/Agent-Field/agentfield/pull/673af install by @AbirAbbas in https://github.com/Agent-Field/agentfield/pull/738Note truncated.
Adopts the mypy half of #620 (the ruff half shipped in #812). mypy.ini existed but nothing ran it. This wires a required mypy step into the Python SDK CI without changing any behavior.
Phase 1 of #971:
The gate is green today (Success: no issues found). It fails if a non-allowlisted module gains a new error, or if a module is removed from the allowlist while it still has errors. No '# type: ignore' sprinkles were added.
Addresses review on #1052. The locked dev environment pins requests 2.33.1, which ships no type stubs, so 'uv sync --extra dev' followed by mypy failed with [import-untyped] in multimodal.py and agent_field_handler.py. CI was green only because 'pip install .[dev]' resolved a newer requests that bundles stubs, so it wasn't validating the reproducible environment.
Pin types-requests==2.33.0.20260906 in the dev extras (and uv.lock) so mypy passes in both the locked (uv sync) and pip-resolved environments. Verified mypy clean against the locked requests 2.33.1 and a newer requests; ruff still clean; uv lock --check clean.
session_turn_detection.py landed on main (via #1056) after this PR opened and carries 3 mypy errors (loosely-typed dict value used in numeric comparisons at line 58). The mypy gate correctly failed on it because it was not allowlisted - the ratchet working as designed.
Add it to the allowlist to keep this PR a pure CI-wiring change; the module's type errors can be cleaned in a follow-up (phase 2). (8c3e084)
Bump vulnerable dependencies across the monorepo to patched releases:
Regenerated affected npm/pnpm lockfiles and go.sum.
Co-authored-by: Santosh kumar santoshkumarradha@users.noreply.github.com
Full npm install (not package-lock-only) so control-plane web client and desktop locks include all transitive deps required by npm ci.
Co-authored-by: Santosh kumar santoshkumarradha@users.noreply.github.com
Vitest 4.1.11, hono 4.13.7, and esbuild 0.28.1 were already patched on this branch for the related alerts.
Co-authored-by: Santosh kumar santoshkumarradha@users.noreply.github.com
Co-authored-by: Cursor Agent cursoragent@cursor.com Co-authored-by: Santosh kumar santoshkumarradha@users.noreply.github.com (44d1022)
security(deps): bump pytest to >=9.0.3 (CVE-2025-71176) by @AbirAbbas in https://github.com/Agent-Field/agentfield/pull/488
⚠️ This is a staging/pre-release version for testing. Not recommended for production use.
# Staging binary (use --staging flag)
curl -fsSL https://agentfield.ai/install.sh | bash -s -- --staging
# Python SDK (prerelease - requires --pre flag)
pip install --pre agentfield
# TypeScript SDK
npm install @agentfield/sdk@next
VERSION=v0.1.139-rc.4 curl -fsSL https://agentfield.ai/install.sh | bash
Download the binary for your platform below, make it executable, and move it to your PATH.
agentfield-darwin-amd64agentfield-darwin-arm64agentfield-linux-amd64agentfield-linux-arm64af call rejecting valid input for optional reasoner params by @AbirAbbas in https://github.com/Agent-Field/agentfield/pull/610_current_status issue where status stuck on `S… by @DebanKsahu in https://github.com/Agent-Field/agentfield/pull/673af install by @AbirAbbas in https://github.com/Agent-Field/agentfield/pull/738Note truncated.
fix(storage): rebind transactional prepares and guard stale retries
test(storage): verify stale reapers against live PostgreSQL
Adds connection-backed PostgreSQL coverage for the transactional prepare rebinding and the paired-execution staleness guard:
Live tests are gated on POSTGRES_TEST_URL, refuse non-loopback hosts, and create a throwaway database per test.
RetryStaleWorkflowExecutions repeated the activity predicate in its workflow UPDATE, but the paired execution UPDATE still only rechecked status. A heartbeat committing between the two statements was overwritten, dragging a live execution back to pending.
Repeat the execution staleness predicate on the second update so activity that lands after the workflow guard has been evaluated is respected. The workflow guard and RowsAffected semantics are unchanged.
Reliability coverage:
RetryStaleWorkflowExecutions committed the workflow half of a candidate even when the paired execution update lost to a heartbeat that landed between the two statements. That left workflow=pending and execution=running with retry_count incremented, and the candidate was reported as retried, so a dispatcher could duplicate live work.
Check the paired execution update's RowsAffected and wrap each candidate in a savepoint. When the execution does not move, roll the candidate's workflow half back so the pair is all-or-nothing and the id is not reported as retried.
Savepoint rollback is used rather than locking both rows first because it behaves identically on SQLite and PostgreSQL (SQLite has no row locks, so a lock-first pair could not be exercised by the SQLite regression), keeps the existing lock ordering unchanged, and lets the pre-existing workflow-only paths keep their outcome: candidates with no paired execution row, a terminal row, or a waiting row are only rolled back when the paired execution is active but no longer stale, i.e. when it actually won the race.
Both between-statements regressions now assert the all-or-nothing outcome on SQLite and live PostgreSQL, and a new SQLite test pins that unpaired workflows are still retried.
Add executable regressions on top of the savepoint-based retry fix without changing the implementation:
security(deps): bump pytest to >=9.0.3 (CVE-2025-71176) by @AbirAbbas in https://github.com/Agent-Field/agentfield/pull/488
⚠️ This is a staging/pre-release version for testing. Not recommended for production use.
# Staging binary (use --staging flag)
curl -fsSL https://agentfield.ai/install.sh | bash -s -- --staging
# Python SDK (prerelease - requires --pre flag)
pip install --pre agentfield
# TypeScript SDK
npm install @agentfield/sdk@next
VERSION=v0.1.139-rc.3 curl -fsSL https://agentfield.ai/install.sh | bash
Download the binary for your platform below, make it executable, and move it to your PATH.
agentfield-darwin-amd64agentfield-darwin-arm64agentfield-linux-amd64agentfield-linux-arm64af call rejecting valid input for optional reasoner params by @AbirAbbas in https://github.com/Agent-Field/agentfield/pull/610_current_status issue where status stuck on `S… by @DebanKsahu in https://github.com/Agent-Field/agentfield/pull/673af install by @AbirAbbas in https://github.com/Agent-Field/agentfield/pull/738Note truncated.
Adds SEARCH_PROVIDER=serply alongside the existing Tavily default and
the Parallel Search opt-in, following the same shape as the Parallel
backend: one normalizer, one async executor, one dispatch branch, and no
new entry in requirements.txt.
Serply answers from Google's result pages and needs an API key, so the backend reads SERPLY_API_KEY and fails with the same message shape the Tavily path already uses when its key is missing.
Two details specific to this API are handled explicitly:
num is an approximate upper bound with a ceiling of 10, so the client
requests the ceiling and trims to the example's own five-result budget
rather than trusting the count it gets back.detail string, so the failure
message carries that text instead of a bare status code. (6d36643)fix(sdk/go): return isolated session definitions
fix(sdk/go): distinguish overlapping slice views (7a1714e)
Nothing published for this version
One column per month.