agent-framework-openai

Added

• agent-framework-core , agent-framework-mistral : Add a Mistral chat client with native chat, streaming, tools, structured output, and embeddings support ( #7392 )

• agent-framework-core : Add experimental AGENT-HOOKS-0.1 enforcement middleware behind the opt-in agent-hooks extra ( #7515 )

• agent-framework-openai : Add request preparation and response parsing hooks to OpenAIChatCompletionClient ( #7028 )

• agent-framework-ag-ui : Add workflow checkpoint creation and resume support to AgentFrameworkWorkflow ( #6646 )

• agent-framework-core : Add BackgroundAgentsProvider.release_session() for safely cancelling work and releasing per-session runtime state ( #7450 )

• agent-framework-foundry-hosting : Add provider-based Foundry state stores for agent sessions, checkpoints, and function approvals ( #7533 )

• agent-framework-foundry : Export the Foundry hosted-agent session-state key used for service session continuity ( #7608 )

• agent-framework-orchestrations : Expose the Magentic orchestrator manager name as MagenticOrchestrator.MANAGER_NAME ( #7350 )

• agent-framework-gemini : Surface Gemini thought summaries as reasoning content ( #7488 )

• samples : Add a locally hosted Responses sample for the agent harness ( #7010 )

Changed

• agent-framework , agent-framework-core , agent-framework-azurefunctions , agent-framework-durabletask : Move the Durable Task and Azure Functions integrations to the Durable Agent Framework extension repository ; core continues to re-export their public symbols and install their independently published packages through the all extra ( #7465 )

• agent-framework-core : [BREAKING — experimental] Require functional workflow definitions to be built into stateful workflow instances before running or adapting them as agents ( #7521 )

• agent-framework-foundry-hosting : [BREAKING — beta] Migrate Foundry Hosted Agents to the Agent Server Responses 2.x storage model ( #7533 )

• agent-framework-foundry-hosting , agent-framework-azure-contentunderstanding : Update Azure Agent Server dependencies to the 2.1 beta line and adapt request and storage handling ( #7621 )

• agent-framework-foundry : Make encrypted reasoning opt-in for Foundry chat requests ( #7536 )

• agent-framework-mem0 : Separate Mem0 storage and search scopes ( #7531 )

• agent-framework-ag-ui : Consolidate thread-snapshot ownership and remove unused internal orchestration helpers ( #7426 , #7479 )

• agent-framework-devui : Update frontend transitive dependencies ( #7493 , #7554 )

• agent-framework-mistral , agent-framework-ollama : Expand the supported uv_build version range ( #7445 )

• tests : Update Python development tools and improve sample validation with deterministic replay ( #7350 , #7445 , #7541 , #7545 )

• samples : Update frontend dependencies and add a Foundry Hosted Agents custom-storage sample ( #7529 , #7621 )

Fixed

• agent-framework-core : Ignore excluded tool results during compaction ( #7391 )

• agent-framework-core : Bound tool-result compaction summaries before provider calls ( #7396 )

• agent-framework-core : Report evaluator results with zero checks correctly ( #7399 )

• agent-framework-core : Reject Windows junctions while discovering and accessing skills ( #7507 )

• agent-framework-core : Warn when advertised MCP archives are rejected ( #7622 )

• agent-framework-core : Prevent streaming transcript duplication with message injection and per-service-call persistence ( #7605 )

• agent-framework-ag-ui , agent-framework-core : Preserve conversation correlation across AG-UI runs ( #7430 )

• agent-framework-ag-ui : Preserve approval resume semantics at the protocol boundary ( #7480 )

• agent-framework-ag-ui : Make approval lifecycle, interruption recovery, and replay behavior occurrence-safe ( #7594 )

• agent-framework-azure-ai-search : Forward query-source identity to Azure AI Search ( #7278 )

• agent-framework-azure-cosmos-memory : Call the renamed Cosmos toolkit registration API ( #7635 )

• agent-framework-claude : Avoid reusing one Claude SDK client across distinct fresh sessions ( #7404 )

• agent-framework-copilotstudio : Handle large activities without line-length failures, update the Copilot Studio client dependency, and declare its required HTTP and authentication runtime dependencies ( #7417 )

• agent-framework-declarative : Preserve falsey EditTableV2 items and robustly extract JSON from declarative workflow responses ( #7380 , #7550 )

• agent-framework-foundry : Preserve Foundry hosted-agent session IDs independently from conversation continuation IDs ( #7608 )

• agent-framework-gemini : Restore thought_signature values during approval replay ( #7546 )

• agent-framework-github-copilot : Scope under-specified approve-for-session decisions to the current tool and align package metadata with the SDK's Python 3.11 minimum ( #7607 )

• agent-framework-mistral : Preserve prompt-cache usage details ( #7597 )

• agent-framework-openai : Prevent orphaned local approval responses and safely ignore non-string Chat Completions content ( #7462 , #7028 )

• agent-framework-redis : Honor a max_messages retention limit of zero ( #7470 )

Full Changelog : python-1.13.0...python-1.14.0

Assets 2 Loading

Uh oh!

There was an error while loading. Please reload this page .

👀 2 jijaraba and mdimai666 reacted with eyes emoji All reactions

• 👀 2 reactions

2 people reacted

Added

• agent-framework-core , agent-framework-foundry-hosting : Add bounded, in-memory archive skill discovery for MCP sources and expose archive controls through FoundryToolbox ( #7121 )

• agent-framework-core , agent-framework-foundry-hosting , agent-framework-hosting , agent-framework-hosting-responses , agent-framework-hosting-telegram : Add reusable session stores and persist complete Foundry Responses sessions ( #7306 )

• agent-framework-a2a , agent-framework-ag-ui , agent-framework-anthropic , agent-framework-azure-ai-search , agent-framework-azure-contentunderstanding , agent-framework-azure-cosmos , agent-framework-azure-cosmos-memory , agent-framework-azurefunctions , agent-framework-bedrock , agent-framework-chatkit , agent-framework-claude , agent-framework-copilotstudio , agent-framework-core , agent-framework-declarative , agent-framework-devui , agent-framework-durabletask , agent-framework-foundry , agent-framework-foundry-hosting , agent-framework-foundry-local , agent-framework-gemini , agent-framework-github-copilot , agent-framework-hosting , agent-framework-hosting-a2a , agent-framework-hosting-mcp , agent-framework-hosting-responses , agent-framework-hosting-telegram , agent-framework-hyperlight , agent-framework-lab , agent-framework-mem0 , agent-framework-mistral , agent-framework-monty , agent-framework-ollama , agent-framework-openai , agent-framework-orchestrations , agent-framework-purview , agent-framework-redis , agent-framework-tools : Add process-wide feature-usage telemetry with first-party User-Agent reporting ( #7420 )

• agent-framework-core , agent-framework-openai : Add OpenAI cache-write token usage details to observability data ( #7369 )

• agent-framework-openai : Support ephemeral per-request instructions in the Responses API ( #7292 )

• agent-framework-foundry-hosting : Support asynchronous credentials in FoundryToolbox ( #7208 )

• samples : Demonstrate archive-backed MCP skills with progressive disclosure in the Foundry Toolbox sample ( #7121 )

• samples : Add FileMemoryProvider, TodoProvider, AgentModeProvider, and GitHub Copilot BYOK samples ( #7309 , #7336 , #7428 )

Changed

• agent-framework-ag-ui , agent-framework-core : [BREAKING] Make workflow checkpoints fully replayable from initial input and human-in-the-loop responses ( #7374 )

• agent-framework-devui : Update frontend transitive dependencies ( #7232 )

• samples : Remove numeric prefixes from Foundry Hosted Agent sample names and update frontend dependencies ( #7314 , #7315 , #7405 )

• tests : Add isolated lower- and upper-bound release validation and improve Python package-management tooling ( #7274 , #7342 )

Fixed

• agent-framework-core : Restrict unpickler module-prefix allowances to type objects ( #5923 )

• agent-framework-core : Apply MCP header_provider headers to initialization and ambient requests ( #7305 )

• agent-framework-core : Enforce the documented target-type check in SerializationMixin.from_dict ( #7256 )

• agent-framework-ag-ui : Preserve model emission order in MESSAGES_SNAPSHOT events ( #7239 )

• agent-framework-foundry : Prevent agent-reference requests from inheriting OPENAI_CHAT_MODEL ( #7283 )

• agent-framework-core : Bound summarization input before provider calls ( #7375 )

• agent-framework-openai : Sanitize Chat Completions message author names ( #7127 )

• agent-framework-anthropic : Prevent streaming token usage from being counted twice ( #7162 )

• agent-framework-core : Preserve declaration-only streaming metadata ( #7409 )

• agent-framework-openai : Preserve approval decisions across Responses API continuations ( #7407 )

• agent-framework-core : Omit tool content returned after function invocation limits are reached ( #7408 )

• agent-framework-core : Keep function-call and result occurrences atomic during compaction ( #7406 )

• agent-framework-ag-ui : Correlate confirm_changes snapshots by call ID ( #7411 )

• agent-framework-ag-ui : Defer provider-injected approvals to in-run execution ( #7410 )

• agent-framework-foundry-hosting : Store hosted checkpoints under the durable home directory ( #7220 )

• agent-framework-ag-ui , agent-framework-core : Improve function approval resume and replay behavior ( #7345 )

• agent-framework-openai : Pass raw JSON Schema response formats through Chat Completions without wrapping ( #7199 )

• agent-framework-core : Extract keywords from non-English text for topic selection ( #7130 )

• agent-framework-gemini : Preserve harness tool declarations ( #7322 )

• agent-framework-core : Handle callable-class middleware safely during type detection ( #7333 )

• agent-framework-core : Reject Windows junctions in FileSystemAgentFileStore ( #7291 )

• agent-framework-azurefunctions , agent-framework-core , agent-framework-durabletask : Preserve sub-workflow state during checkpoint restoration ( #7097 )

• agent-framework-core : Prevent duplicate function calls after approval round trips ( #7271 )

• agent-framework-foundry-hosting : Preserve authentication credentials across FoundryToolbox reconnections ( #7202 )

• agent-framework-ag-ui : Scope workflow caches correctly ( #7277 )

• samples : Fix stale agent.json references in the A2A sample ( #7281 )

• tests : Fix GitHub Actions token propagation for Python integration and sample-validation workflows ( #7427 )

Assets 2 Loading

Uh oh!

There was an error while loading. Please reload this page .

👍 2 mdimai666 and jiachangkun02 reacted with thumbs up emoji All reactions

• 👍 2 reactions

2 people reacted

Added

• agent-framework-azure-cosmos-memory : Add an alpha Azure Cosmos DB semantic-memory context provider with fact extraction, user profiles, samples, and integration coverage ( #6719 )

• agent-framework-azurefunctions , agent-framework-core , agent-framework-durabletask : Add HITL response-URL addressing for requests raised from inside workflows ( #7001 )

• agent-framework-core : Add cross-session origin attribution to context-injected messages ( #7041 )

• agent-framework-core , agent-framework-tools : Warn when auto-approved tools have name collisions ( #7090 )

• agent-framework-core : Add a session_provider option to MCPSkillsSource and MCPSkill (mutually exclusive with client ) that resolves the MCP session on every fetch, keeping cached skills reconnect-safe when the underlying session is replaced ( #7135 )

• agent-framework-hosting-a2a : Add app-owned A2A hosting helpers ( #7050 )

• agent-framework-hosting-mcp : Add app-owned MCP hosting helpers for exposing agents and workflows as native MCP tools ( #7209 )

• agent-framework-hosting-responses : [BREAKING] Add Responses conversation ID creation and parsing helpers, and distinguish conversation IDs from previous response IDs ( #7234 )

• agent-framework-hosting-telegram : Add Telegram hosting helpers and samples ( #7047 )

• samples : Add a Microsoft OpenTelemetry Distro observability sample ( #5632 )

Changed

• agent-framework-ag-ui : [BREAKING] Emit TOOL_CALL events for workflow participant tool calls ( #7039 )

• agent-framework-a2a : Reduce A2AExecutor log noise for content types without protocol mappings ( #7034 )

• agent-framework-ag-ui , agent-framework-core : Optimize shared serialization paths ( #7165 )

• agent-framework-ag-ui , agent-framework-bedrock , agent-framework-claude , agent-framework-core , agent-framework-github-copilot , agent-framework-ollama , agent-framework-openai : Normalize chat finish reasons across providers ( #7105 )

• agent-framework-anthropic , agent-framework-azure-contentunderstanding , agent-framework-azure-cosmos , agent-framework-core , agent-framework-declarative , agent-framework-devui , agent-framework-foundry , agent-framework : Update Microsoft Foundry branding in shipped APIs and package documentation ( #6999 )

• agent-framework-azure-contentunderstanding , agent-framework-azure-cosmos-memory , agent-framework-chatkit , agent-framework-core , agent-framework-durabletask , agent-framework-foundry , agent-framework-foundry-hosting , agent-framework-gemini , agent-framework-hyperlight , agent-framework-lab , agent-framework-monty , agent-framework-openai , agent-framework-tools , agent-framework : Consolidate dependency updates and compatibility adjustments ( #7204 )

• agent-framework-azurefunctions , agent-framework-devui , agent-framework-foundry , agent-framework-hosting-responses , agent-framework-lab , agent-framework-mistral , agent-framework : Clean up dependency groups and compatibility handling ( #7046 )

• agent-framework-azurefunctions , agent-framework-durabletask : Normalize initial durable workflow inputs across hosting paths ( #7205 )

• agent-framework-core : [BREAKING — experimental] Correct harness before-strategy compaction when state persists per service call ( #7055 )

• agent-framework-core : [BREAKING] Graduate create_harness_agent from experimental to stable ( #7120 )

• agent-framework-core : Graduate the mode and todo providers from experimental to stable ( #7053 )

• agent-framework-core : Graduate ToolApprovalMiddleware from experimental to stable ( #7106 )

• agent-framework-core : Graduate FileMemoryProvider from experimental to stable ( #7113 )

• agent-framework-core : Make FileAccessProvider opt-in for harness agents ( #7094 )

• agent-framework-core : Serialize tool definitions best-effort for observability ( #7029 )

• agent-framework-declarative : Promote declarative workflows from release candidate to stable ( #7065 )

• agent-framework-devui : Refine request logging ( #7083 )

• agent-framework-foundry-hosting : Promote the package to beta and add it to the main installation surface; make the Foundry Toolbox MCP skills sample self-contained ( #7099 )

• agent-framework-azure-contentunderstanding , agent-framework-gemini , agent-framework-mistral , agent-framework-monty , agent-framework-tools : Promote the packages to beta, add them to the main installation surface, expose lazy-loading namespaces, and move package-local samples into the root sample tree

• agent-framework-github-copilot : Forward GitHubCopilotOptions verbatim when creating sessions ( #7155 )

• docs : Add self-hosting sample snippets ( #7104 )

• docs : Add environment-file templates for Durable Task hosting samples ( #5948 )

• samples : Keep ChatKit attachments close to the sample application that owns them ( #7038 )

Fixed

• agent-framework-ag-ui : Bind streamed tool arguments to their call ids ( #6342 )

• agent-framework-ag-ui : Accept state data URIs whose media type includes parameters ( #6905 )

• agent-framework-ag-ui : Coalesce reasoning deltas without content ids into a single reasoning block ( #6804 )

• agent-framework-ag-ui : Bridge request state and session continuity ( #7084 )

• agent-framework-ag-ui : Replay workflow handoff results correctly ( #7102 )

• agent-framework-ag-ui : Clarify require_confirmation documentation for confirm_changes HITL gating ( #6884 )

• agent-framework-anthropic : Prevent per-run additional_beta_flags from leaking into request keyword arguments ( #7060 )

• agent-framework-core : Clear service_session_id in the agent wrapper when session propagation is enabled ( #5875 )

• agent-framework-core : Preserve tool span context for parallel calls ( #6512 )

• agent-framework-core : Parse structured values assembled from split text chunks ( #6990 )

• agent-framework-core : Raise ValueError for malformed data URIs ( #6916 )

• agent-framework-core : Preserve function-call names when merging streaming deltas ( #6809 )

• agent-framework-core , agent-framework-durabletask : Handle checkpoint encodings consistently ( #6579 )

• agent-framework-core : Preserve explicit null arguments during automatic function calling ( #7108 )

• agent-framework-core : Count non-ASCII text correctly during compaction ( #7124 )

• agent-framework-core : Forward header_provider headers to streamable HTTP MCP transports ( #7218 )

• agent-framework-core : Prevent compaction from emitting empty projections ( #7219 )

• agent-framework-core : Return MCP tool-use sampling results to the requesting server ( #7189 )

• agent-framework-foundry-hosting : Make FoundryToolbox.as_skills_provider() cache toolbox skill discovery by default so skill://index.json is read once instead of on every agent run, give disable_caching an observable effect, and add a cache_refresh_interval o...

Read more

Assets 2 Loading

Uh oh!

There was an error while loading. Please reload this page .

All reactions

Previous 1 2 3 4 5 6 7 Next

Previous Next

You can’t perform that action at this time.

Added

• agent-framework-core : Add message injection middleware so tools or host code can enqueue messages into an active run and drain them into the next model call within the same AgentSession ( #6998 )

• agent-framework-core : Integrate message injection into create_harness_agent and the harness console sample so a running harness agent can be nudged mid-turn ( #7027 )

• agent-framework-core : Add progressive MCP disclosure so agents can discover, load, and unload MCP tool schemas on demand while keeping the allowed_tools boundary intact ( #6850 )

• agent-framework-core : Add refresh_interval (TTL) to CachingSkillsSource so cached skill lists expire and re-fetch on a configured interval ( #6977 )

• agent-framework-core , agent-framework-foundry-hosting : Add SkillsSourceContext (invoking agent plus optional session) threaded through the skills source pipeline, enabling context-aware filtering and per-key cache isolation ( #6895 )

• agent-framework-core : Allow disabling approval for SkillsProvider tools ( #6867 )

• agent-framework-core : Allow opting out of FileAccessProvider tool approval ( #6879 )

• agent-framework-core : Allow custom argument parsing for inline skill scripts so non-conforming tool-call argument shapes (for example, vLLM) can be handled ( #6817 )

• agent-framework-hosting , agent-framework-hosting-responses : Add a hosting protocol helper surface ( AgentState , WorkflowState , SessionStore , AgentRunArgs , WorkflowRunArgs ) and Responses helpers ( create_response_id , responses_session_id , responses_to_run , responses_from_run , responses_from_streaming_run ) ( #6891 )

• agent-framework-ag-ui : Add FastAPI SSE keepalive support for long, output-silent streams ( #6980 )

• agent-framework-github-copilot : Forward skill_directories and disabled_skills to the Copilot session ( #6937 )

• agent-framework-openai : Allow tool_choice: required when allowed_tools is set ( #7024 )

• agent-framework-anthropic , agent-framework-core , agent-framework-foundry-hosting , agent-framework-gemini , agent-framework-openai : Mark hosted/provider-executed tool calls as informational-only via Content.informational_only so they remain visible in transcripts without local re-invocation ( #6997 )

• samples : Add a deterministic action-boundary validation middleware sample ( #6528 )

• samples : Add Agent Harness blog post accompanying samples, part 3 ( #6741 )

• samples : Add a declarative Foundry Hosted Agent workflow sample ( #6897 )

Changed

• agent-framework-core : [BREAKING — experimental] Extract caching from SkillsProvider into a CachingSkillsSource decorator ( #6847 )

• agent-framework-core : [BREAKING — experimental] Treat nested SKILL.md content as part of the parent skill instead of discovering it as a separate skill root ( #6849 )

• agent-framework-core : [BREAKING — experimental] FileAccess / FileMemory replace_lines now performs literal replacement (including line deletion) instead of always re-adding a line terminator ( #6859 )

• agent-framework-core : Remove the experimental marker from the Skills API now that its surface is stable ( #6974 )

• agent-framework-core : Lazy-load root agent_framework exports to reduce import cost for narrow-surface scenarios ( #6962 )

• agent-framework-a2a , agent-framework-claude , agent-framework-copilotstudio , agent-framework-core , agent-framework-durabletask , agent-framework-github-copilot , agent-framework-purview : Implement ADR-0029 service_session_id lifecycle mapping, separating durable continuation state, per-run identity forwarding, and telemetry conversation-id extraction ( #6724 )

• agent-framework-azurefunctions , agent-framework-core , agent-framework-durabletask : [BREAKING] Support multi-workflow hosting and sub-workflows on the Durable Task host, including per-workflow durable naming and nested human-in-the-loop request routing ( #6696 )

• agent-framework-ag-ui : [BREAKING] Canonicalize AG-UI interrupt and resume handling around RUN_FINISHED.outcome.interrupts and canonical ResumeEntry payloads ( #6925 )

• agent-framework-mem0 : Support the mem0ai 2.x OSS search call shape ( #7004 )

• agent-framework-mistral : Widen the uv_build backend requirement to allow newer uv releases ( #7033 )

• agent-framework-lab : Raise the agentlightning dependency ceiling for the lightning extra ( #6984 )

• agent-framework-claude , agent-framework-durabletask , agent-framework-gemini , agent-framework-monty , agent-framework-openai : Raise dependency floors to the first versions that provide the SDK APIs and typing consumed by these packages

• docs : Clarify AgentSession.service_session_id scoping to document backing API key/project boundaries and hosted multi-tenant guidance ( #6993 )

• docs : Add security guidance for external skill sources and script execution to harness feature docstrings ( #6936 )

• samples : Bump vite and @vitejs/plugin-react-swc in the ChatKit integration sample frontend ( #6613 )

• samples : Add a multi-tenant hosting security consideration note to the A2A sample ( #6983 )

• samples : Update Foundry Hosted Agent samples for the v2 protocol changes ( #6841 )

• samples : Use a writable runtime directory for the Foundry Skills sample ( #6606 )

• tests : Add Agent typing smoke tests across chat clients ( #6950 )

• tests : Skip NumPy stubs during mypy typing to unblock scheduled dependency-maintenance typing runs ( #6969 )

• tests : Consolidate Dependabot dependency updates for dev tooling ( uv , ruff , pytest , mypy , pyright , mcp , opentelemetry-sdk , poethepoet ) across the workspace and package dev-dependency groups ( #6984 , #7033 )

• tests : Bump the transitive js-yaml dependency in the DevUI frontend lockfile ( #6813 )

Fixed

• agent-framework-core : Parse the structured response value from the final message instead of concatenated text, avoiding spurious ValidationError / JSONDecodeError ( #6383 )

• agent-framework-core : Fix read_skill_resource instruction dropping the .md extension ( #7031 )

• agent-framework-core : Bind policy-enforcement approvals to a single tool invocation (call id, function, arguments, security label, and session) and consume them on first use ( #6966 )

• agent-framework-core : Process messages to an executor serially within a superstep to prevent concurrent handler invocations for the same target executor ( #6776 )

• agent-framework-core : Auto-inject local conversation history on stateless clients even when non-history context providers (for example, SkillsProvider and FileAccessProvider ) are registered ( #6810 )

• agent-framework-core : Improve the error message when a TypeVar is used in handler/executor registration ( #4553 )

• agent-framework-anthropic , agent-framework-core : Fix Anthropic requests that mix tool calls and tool results in one assistant message, and return a deterministic result when the function-loop limit is reached with a blank final response ( #6794 )

• agent-framework-anthropic , agent-framework-core , agent-framework-foundry-hosting , agent-framework-openai : Fix Foundry reasoning/MCP compaction so reasoning output keeps its provider id and ...

Read more

Assets 2 Loading

Uh oh!

There was an error while loading. Please reload this page .

👍 2 Yuya218 and ghominejad reacted with thumbs up emoji All reactions

• 👍 2 reactions

2 people reacted

Previous 1 2 3 4 5 6 7 Next

Previous Next

You can’t perform that action at this time.