v3.2.02026-11-12
Summary
Agno v3.2 makes Super Agent faster and more capable with direct tools, expands Company task governance and coordination, introduces database-backed AgentOS homepage metrics, upgrades A2A to the v1.0 protocol, updates the Slack interface to use the Bolt SDK and makes external agents first-class stored components.
Changelog
Breaking changes
- OS Manager defaults and validation: OSManager.toolcalllimit now defaults to 24 (was 8), and numhistoryruns defaults to 5 (was 7). AgentOS refuses startup when OSManager(root=...) is combined with custom tools named readfile, listfiles, or search_content.
- A2A v1.0: AgentOS and A2AClient now run on the official a2a-sdk 1.x, so upgrade the Agno server and client together. (#10956)
- The /v1/message: and /v1/tasks: routes are removed; use each entity's A2A endpoint instead.
- Stream events and response artifacts follow the v1.0 shape.
- The a2a extra now requires a2a-sdk[http-server]>=1.2.2,<2.
- Direct imports of removed A2A helpers need updating.
- Standard v0.3 clients still work through the compatibility wire by default.
- Slack interface: the interface now runs on Slack Bolt, and only the person who started a run can Approve or Deny it, unless they're named in approvers= or mapped to the same user with users=. (#11599)
- The slack extra now requires slackbolt>=1.29.0,<2 and slacksdk>=3.43.0,<4; run pip install -U "agno[slack]".
- List anyone else who approves runs in approvers=.
- AgentOS refuses startup when Slack serves agent_os.manager under AgentOS(authorization=...) or an authorization middleware without users=.
- Direct imports of removed internal Slack helpers (SlackEventHandler, HITLHandler, verifyslacksignature) need updating.
New features
- Super Agent: Super Agent now does the work itself instead of routing requests through manager members. Platform Builder is its only member. (#11572)
- Lists and runs agents, teams, and workflows, and reads platform activity and schedules.
- Uses Company, docs, source-file, and custom OS Manager tools, and reads the current chat history.
- Its context includes accessible components, and every tool keeps its own permission check.
- Company OS: Routines get a dedicated runner, Tasks carry richer state, and published Studio components can execute work.
- Mini Agent for Routines: Routines run on a restricted, non-chat Mini Agent with Company tools and packaged Routine skills, but no platform or general run tools. It is hidden from agent discovery, MCP, and A2A, and refuses non-Routine runs. (#11572)
- Task coordination: Tasks expose lastrun and API-computed allowedtransitions, support direct start, and record durable attempt state. Super Agent can start an accepted Task on the durable queue without blocking the conversation; human acceptance and review are unchanged. (#11430, #11562)
- Company details: set a Company name and description in config, through /company/details, or from chat, with confirmation before replacing existing details. Super Agent gets fresh Company context on every run plus concise coordination guidance. (#11502, #11569, #11563)
- Studio components as executors: published Studio agents, teams, and workflows can execute Company Tasks and Routines. Drafts cannot, and normal executor safety checks still apply. (#11443)
- AgentOS homepage metrics: new component, aggregate, and refresh tables back daily sessions, runs, tokens, users, models, cost, duration, time-to-first-token, and model latency through /os/metrics/*. Refreshes run in the background on all supported databases, rebuild only changed days, support ETags, and aggregate across multiple local or remote databases. (#10874)
- Stored external-agent components: ClaudeAgent and CodexAgent can be saved, loaded, versioned, listed, and run through the components API. Sandbox and output-schema references resolve through the registry, and unsupported or non-serializable settings fail explicitly instead of being silently dropped. (#11268)
- A2A stored tasks and lifecycle: A2A tasks are stored in the database and support send, stream, get, cancel, list, and subscribe. External agents are served, HITL and external tools can pause and continue, and Agent Cards advertise both v1.0 and v0.3 interfaces. (#10956)
- Session-run pagination: GET /sessions/{sessionid}/runs supports opt-in cursor pagination via limit, beforerunindex, and afterrun_index. Pages keep top-level runs together with their nested member and step runs, and calls without limit still return a bare list. (#8456)
- Slack interface: Slack now supports the agent view, Stop, external agents and Super Agent results. Assistant-view apps keep working. (#11599)
- Agent view: the shipped manifest turns on Slack's agent view, with suggested prompts on the Messages tab and a thread title set from the first direct message.
- Streaming: replies stream with task cards and a status line, and long replies or runs past nine minutes continue in a new message instead of being cut.
- Stop: pressing Stop cancels the run for Agents, Teams, Workflows, external agents and remote entities.
- External agents: ClaudeAgent and CodexAgent run behind the interface, and their native approvals are answered from Slack, including sandboxed agents on the durable queue.
- Super Agent delivery: Slack(team=agent_os.manager) serves Super Agent, and a Company Task it starts posts its result once into the thread that started it, across restarts.
- User mapping: users= maps Slack users to AgentOS principals so each message runs with that principal's roles, and approvers= names who may approve, deny or stop any run.
Improvements
- Studio component ID recovery: Studio tools normalize surrounding whitespace and invisible format characters in component IDs. Read-only get_component can resolve a visible leading ID followed by stray text and warns with the exact ID; mutating tools remain strict. (#11572)
- Sandbox operation and deployment: Redis is optional for sandboxed agents on the durable queue; Postgres plus a shared checkpoint store now handles cross-replica cancellation, resume, and recovery. Checkpoint restore/capture also run outside long-lived database transactions, avoiding blocked writers, exhausted connections, and lease expiry during large transfers. (#11217, #11192)
- Metrics accuracy and compatibility: run/session token totals now avoid double counting, include failed/cancelled external runs and all model/tool-loop calls, and report corrected timing and provider usage. MongoDB, Valkey, and DynamoDB legacy-run refreshes are corrected and optimized; no-op refresh messages move to debug; fresh installs no longer create the deprecated agno_metrics table. (#11426, #11574, [#11570](https://github.com/ag
v3.1.122026-10-29
Changelog
New Features:
- Native Approvals for Harness Agents: Claude's and Codex's own permission requests now reach the AgentOS approval UI. When the agent has a database, including one AgentOS supplies, a native request persists a PAUSED run that shows up in the existing chat and Approvals views, and Resolve plus Continue Run delivers the decision back to the same native execution without replaying the prompt. The harness's own policy decides when to ask — Claude keeps its permissionmode and explicit permission callbacks, Codex supports approvalpolicy="on-request" | "untrusted" | "never" alongside its existing SDK autoreview and denyall modes — so there is no separate Agno setting to turn on. Decisions are validated against the exact action requested, with sync and async continuation, cancellation, expiry and storage-failure handling, and sandboxed agents keep this across Docker and AgentCore: capability negotiation, a decision RPC, fenced persistence, reattachment after an API-worker loss, separate approval-wait and execution budgets, and no replay after process loss. Note that a CodexAgent with no approval settings now runs under the native on-request policy, so a request that previously went nowhere pauses the run instead; hitltimeout (default 900s) bounds how long it waits. See cookbook/frameworks/humanintheloop/.
- Initially Disabled Company Routines: Routine(enabled=False) lets a template ship an inactive Routine that a user turns on later from the UI, with True still the default. Disabled Routines are created atomically, stay linked to their Product and visible in listings, are skipped by the scheduler, and keep the user's saved choice across restarts. The flag is honored in Python and YAML bootstrap, the setup and creation APIs, the sync and async Company Manager, and the Company tools. Duplicate create requests still return 409 and cannot enable an existing Routine, and the existing pause/resume endpoints enable it without recreating it. See cookbook/05agentos/30company/11disabled_routine.py.
- Headline Numbers on Charts: ChartTools.create_chart() takes an optional preformatted value — a total or latest value such as "$22.3K" or "94%" — emitted as "value" in the chart block and rendered by Agno OS beside the chart title. The docstring tells the model to leave it out when no single number summarizes the chart, so averages and rates are not summed.
Improvements:
- Company Routine Sorting: GET /company/routines now also sorts by nextrunat, schedule_type and enabled, so Agno OS can sort its routines table by Next run at, Schedule and Active. Paused and on-demand Routines have no upcoming run, so they come last in either direction, and Routines with the same value keep their creation order.
- Cookbooks: Added an ApifyTools job search example that caps the Actor's item count with a tool hook and trims each dataset item to title, company, location, posted date and link, so a broad search stays small in context. Added a fully local Pixeltable RAG example that writes chunks through langchain-pixeltable and searches them with LangChainVectorDb, using local sentence-transformers embeddings and Ollama — no embedding API key and no separate vector database server.
Bug Fixes:
- FastAPI Native Tracing: FastAPI 0.142.0 enables its own OpenTelemetry tracing as soon as a global tracer provider exists, which AgentOS(tracing=True) installs, so FastAPI's request spans were stored alongside Agno's. Every run became a child of the request span and traces were recorded as fastapi.dependencies with empty runid, sessionid and agentid, breaking the Runs view, while each UI request stored a trace of its own. AgentOS now opts the app it builds out of FastAPI's tracing, and setuptracing() filters foreign spans so a stored trace is rooted at the Agno run. Apps without tracing=True keep FastAPI's defaults, and users' own spans created inside tools are still kept.
- Company Routine Cancel: POST /company/routines/{id}/cancel recorded the run as cancelled as soon as the executor accepted the request, but the agent only stops at its next checkpoint. For that window the API reported cancelled while the agent was still working, and archive, delete and executor change went through mid-run. The route now records a status only from durable evidence: a live run is left alone for ScheduleExecutor to settle, a finished run is written through the scheduler's own mapping, a missing run is recorded cancelled, and a crashed run past its timeout no longer blocks archive, delete or executor change.
- AccuracyEval Judge: AccuracyEval with no model and no evaluatoragent built its judge with OpenAIChat(id="o4-mini"), which OpenAI shut down on 2026-10-23, so every judge call returned 404 modelnot_found, no iteration scored, and the eval reported an empty run instead of the provider error behind it. The default is now gpt-5.6-terra, the replacement OpenAI lists for o4-mini. The examples that pinned retired models move across too, and the accuracy examples switch to OpenAIResponses because OpenAIChat rejects function tools for these models.
- Knowledge Per-Upload Chunking: POST /knowledge/content accepts per-upload chunker, chunksize and chunkoverlap, and process_content() applied them straight to content.reader — the shared custom or already-resolved reader on the knowledge base. Every later upload using that reader kept the previous upload's chunker until the process restarted. The upload now gets its own shallow copy of the reader before the chunking strategy is applied, so the knowledge base keeps its own configuration.
What's Changed
* fix: knowledge keep per-upload chunker options off the knowledge base reader by @sunhouy in https://github.com/agno-agi/agno/pull/11327
* fix: move the AccuracyEval judge and examples off retired OpenAI models by @kausmeows in https://github.com/agno-agi/agno/pull/11347
* [cookbook] Add a local Pixeltable RAG example by @pierrebrunelle in https://github.com/agno-agi/agno/pull/11314
* [cookbook] ApifyTools job search with capped runs and trimmed results by @johnisanerd in https://github.com/agno-agi/agno/pull/11305
* feat: show a headline number on ChartTools charts by @anuragts in https://github.com/agno-agi/agno/pull/11110
* fix: keep FastAPI native spans out of Agno traces by @kausmeows in https://github.com/agno-agi/agno/pull/11346
* fix: record a Routine cancel only once its run has stopped [SDK-643] by @harshsinha03 in https://github.com/agno-agi/agno/pull/11352
* chore: README by @ashpreetbedi in https://github.com/agno-agi/agno/pull/11380
* feat: support initially disabled Company routines by @ashpreetbedi in https://github.com/agno-agi/agno/pull/11372
* feat: route native Claude and Codex approvals through AgentOS by @ashpreetbedi in https://github.com/agno-agi/agno/pull/11374
* feat: sort routines by next run, schedule type and enabled by @Ayush0054 in https://github.com/agno-agi/agno/pull/11382
* chore: core release v3.1.12 by @kausmeows in https://github.com/agno-agi/agno/pull/11383
New Contributors
* @sunhouy made their first contribution in https://github.com/agno-agi/agno/pull/11327
* @pierrebrunelle made their first contribution in https://github.com/agno-agi/agno/pull/11314
* @johnisanerd made their first contribution in https://github.com/agno-agi/agno/pull/11305
Full Changelog: https://github.com/agno-agi/agno/compare/v3.1.11...v3.1.12
v3.1.112026-10-27
Changelog
New Features:
- Harness Agents in Teams and Workflows: ClaudeAgent and CodexAgent can now be members of an Agno Team and steps of a Workflow, including inside Parallel, Loop, Condition and Router. The team leader delegates to them with delegatetasktomember like any other member, each harness runs with its own model, tools and workspace, keeps its own conversation across turns of the same session, and the team or workflow collects its runs as usual. A harness agent with sandbox= runs each turn in its own container as a member or step, and AgentOS lists harness members on the team page. See cookbook/frameworks/harnessteams/ and cookbook/frameworks/harness_workflows/.
- Session Search: Added GET /sessions/filter-schema and POST /sessions/search with nested AND/OR/NOT filters on session name, session/user/component IDs, session type and created/updated timestamps, matching the Traces API. Predicates are applied before counting and pagination in PostgreSQL and SQLite, with a batched fallback for other adapters. Sync and async client methods are included.
- Company Tasks: POST /company/tasks accepts status, so Tasks can be created in Backlog, Ready or Cancelled. Tasks can move between any stages, Done and Cancelled Tasks can reopen, and title and brief can be edited at any stage. In progress still starts only with a run, and moving out of it while that run is executing or paused is refused.
- Company Routines: Added POST /company/routines/{id}/cancel, /archive, /restore and admin-only DELETE /company/routines/{id}. GET /company/routines takes sortby and sortorder. Pause and resume on an on-demand Routine return 409, and scheduler runs no longer move updated_at.
Improvements:
- Cookbooks: Added a customer feedback workflow for the AgentOS UI, where Company Manager reads Markdown feedback, proposes evidence-backed Projects and Backlog Tasks, and leaves acceptance to a human.
- README: Aligned the README with the docs introduction and the self-improving agent platform positioning.
Bug Fixes:
- ReaderFactory: createreader() now returns a fresh reader configured with the per-call options instead of a cached singleton, so chunksize and other overrides are respected and mutations no longer leak across Knowledge instances.
- Workflow Step Content: StepOutput.to_dict() keeps ints, floats, bools and strs as themselves instead of stringifying them, so a step that returned 0 or False no longer comes back as "0" or "False" after a HITL resume, a session reload, or through the AgentOS workflow API.
- HuggingFace: defaultheaders is mapped to the SDK's headers, store is sent in extrabody, and maxretries and defaultquery are deprecated and ignored instead of raising TypeError. An agent with no tools no longer sends tools=[], which made non-streaming runs return None content.
- Google Calendar: findavailableslots now works in the authenticated user's timezone instead of UTC, so working hours and date-only inputs mean the user's local day. It also fetches events for the full scanned window, so busy time on the last day of a range is no longer offered as free.
- Company Archive: Archived Tasks no longer block archiving their Project or Product. Visible active Tasks still block as before.
- RedisDb: keyword_search now passes the caller's limit to redisvl instead of silently returning its default of 10 results.
- MMRReranker: rerank and arerank return no documents when limit is zero or negative instead of always returning the closest match.
- FieldLabeledCSVReader: page_size=0 is rejected alongside negative values instead of silently returning no documents on larger files.
- AgentOSClient: Removed the documented but non-existent stream argument from continueagentrun. Use continueagentrun_stream to stream.
What's Changed
* cookbook: add customer feedback UI workflow by @ashpreetbedi in https://github.com/agno-agi/agno/pull/11271
* chore: refresh platform README by @ashpreetbedi in https://github.com/agno-agi/agno/pull/11291
* chore: clarify README self-improvement workflow by @ashpreetbedi in https://github.com/agno-agi/agno/pull/11293
* [fix] Return no documents when the MMR reranker limit is zero by @makiaveli1 in https://github.com/agno-agi/agno/pull/10450
fix: drop the non-existent stream arg from continueagent_run by @yetuge in https://github.com/agno-agi/agno/pull/10453
* chore: align README with self-improving agent platform positioning by @ashpreetbedi in https://github.com/agno-agi/agno/pull/11301
fix: pass the caller's limit to redisvl in RedisDb.keywordsearch by @Ayoubhm07 in https://github.com/agno-agi/agno/pull/10781
* fix: don't let archived Tasks block Company archive by @Ayush0054 in https://github.com/agno-agi/agno/pull/11292
fix: query the scanned window in findavailable_slots by @devYRPauli in https://github.com/agno-agi/agno/pull/10706
* fix: keep JSON-native step content types across a workflow session reload by @kausmeows in https://github.com/agno-agi/agno/pull/11304
* fix: send HuggingFace client and request params under the names the SDK takes by @kausmeows in https://github.com/agno-agi/agno/pull/11307
* [fix] Reject zero page size in field-labeled CSV reader by @jiangjiang248 in https://github.com/agno-agi/agno/pull/10602
* chore: align README with docs introduction by @ashpreetbedi in https://github.com/agno-agi/agno/pull/11313
* fix: use the authenticated user's timezone for calendar slots by @Himanshu040604 in https://github.com/agno-agi/agno/pull/11306
* fix: ensure ReaderFactory respects per-call options and avoids mutable shared state by @Fazlul105 in https://github.com/agno-agi/agno/pull/11296
* test: move the OpenAI reasoning tests off the retired o3-mini by @kausmeows in https://github.com/agno-agi/agno/pull/11331
* test: move the team event streaming tests off the retired o3-mini by @harshsinha03 in https://github.com/agno-agi/agno/pull/11335
* feat: let Claude Code and Codex agents be members of an Agno team by @kausmeows in https://github.com/agno-agi/agno/pull/11241
* feat: add advanced session search and filter schema by @pritipsingh in https://github.com/agno-agi/agno/pull/11319
* feat: Company Task stages on create and move, Routine cancel, archive and delete by @harshsinha03 in https://github.com/agno-agi/agno/pull/11312
* fix: don't let archived Tasks block Project completion or retention by @ysolanky in https://github.com/agno-agi/agno/pull/11336
* chore: Release v3.1.11 by @kausmeows in https://github.com/agno-agi/agno/pull/11334
New Contributors
* @Ayoubhm07 made their first contribution in https://github.com/agno-agi/agno/pull/10781
Full Changelog: https://github.com/agno-agi/agno/compare/v3.1.10...v3.1.11
v3.1.102026-10-23
Changelog
Bug Fixes:
- CodexAgent: Now signs in automatically with OPENAIAPIKEY, and keeps an existing ChatGPT or other login.
- OS Manager: OS Manager now lists and runs external agents such as ClaudeAgent and CodexAgent registered in AgentOS(agents=[...]), by name or ID.
- Knowledge Upload: POST /knowledge/content now returns a clear 400 message when an upload has no file, text_content or URL, or when url is not a single URL.
- Tracing: Spans from AgentOS(tracing=True) and setuptracing() now keep up to 1,024 attributes, so model spans keep their full input and message history on runs with many tool calls. OTELSPANATTRIBUTECOUNTLIMIT and OTELATTRIBUTECOUNTLIMIT still take precedence.
What's Changed
* chore: log replica-local queue stream notice at info level by @ysolanky in https://github.com/agno-agi/agno/pull/11264
* fix: reword the no-authorization user isolation warning by @SamJupe in https://github.com/agno-agi/agno/pull/11263
fix: authenticate Codex automatically with OPENAIAPI_KEY by @ashpreetbedi in https://github.com/agno-agi/agno/pull/11249
* fix: let OS Manager discover and run external agents by @ashpreetbedi in https://github.com/agno-agi/agno/pull/11247
* fix: clarify accepted Company Task execution by @ashpreetbedi in https://github.com/agno-agi/agno/pull/11261
* fix: reject knowledge uploads without a content source by @ashpreetbedi in https://github.com/agno-agi/agno/pull/11259
* fix: clarify readable Company sources for agents by @ashpreetbedi in https://github.com/agno-agi/agno/pull/11256
* fix: reject unsupported knowledge URL input types by @ashpreetbedi in https://github.com/agno-agi/agno/pull/11260
* fix: use LLM-appropriate tracing attribute limits by @ashpreetbedi in https://github.com/agno-agi/agno/pull/11255
* fix: keep Company Task projection off the queue's critical path by @ysolanky in https://github.com/agno-agi/agno/pull/11188
* fix: allow authorized members to trigger Company routines by @ashpreetbedi in https://github.com/agno-agi/agno/pull/11254
* feat: govern Company writes by scope instead of ownership by @SamJupe in https://github.com/agno-agi/agno/pull/11270
* chore: Release v3.1.10 by @harshsinha03 in https://github.com/agno-agi/agno/pull/11265
Full Changelog: https://github.com/agno-agi/agno/compare/v3.1.9...v3.1.10
v3.1.92026-10-22
Changelog
New Features:
- MCP Tool Runner: POST /mcp/server/tools/{toolname}/run runs any tool listed on the AgentOS Server Card over HTTP and returns its result, so a tool can be tested without an MCP client. Turn it off with MCPConfig(toolrun_endpoint=False). See cookbook.
Bug Fixes:
- AgentOS MCP Server: /mcp and /mcp/server-card now accept browser requests from origins the AgentOS CORS policy allows, so os.agno.com can reach the MCP server of a local AgentOS. /info now reports the configured MCP path instead of always /mcp.
- Postgres Metrics: Metrics refresh on a migrated database no longer loads the legacy sessions.runs column into memory.
- Function Workflows: A function passed as a workflow's steps now runs once when its body raises a TypeError, and sync functions and generators now work with arun().
- AgenticChunking: Chunking no longer hangs when the model returns a zero or negative breakpoint, and maxchunksize must be greater than 0.
- BaiduSearchTools: Chinese titles and abstracts are returned as readable text instead of \uXXXX escapes.
- Human-in-the-Loop User Input: A requiresuserinput tool with a parameter typed int | None or another X | Y type now pauses cleanly, and the paused run is saved and resumes with continuerun(runid=...).
- Mistral: The timeout, endpoint and safemode parameters on MistralChat, and endpoint on MistralEmbedder, now work instead of raising a TypeError. safemode is deprecated in favor of safe_prompt.
What's Changed
* [fix] Keep Chinese text readable in BaiduSearchTools output by @RaycarlLei in https://github.com/agno-agi/agno/pull/11229
fix: handle non-positive breakpoints and validate maxchunk_size in AgenticChunking by @Fazlul105 in https://github.com/agno-agi/agno/pull/11231
* fix: run custom workflow steps function once when its body raises TypeError by @kausmeows in https://github.com/agno-agi/agno/pull/11235
* fix: stop Postgres metrics refresh from loading the legacy runs column into memory by @sannya-singal in https://github.com/agno-agi/agno/pull/11190
* feat: run Company automation hourly instead of every minute by @ysolanky in https://github.com/agno-agi/agno/pull/11221
* fix: admit CORS-allowed browser origins on the MCP surface by @ysolanky in https://github.com/agno-agi/agno/pull/11246
* feat: run a published MCP tool over HTTP by @sannya-singal in https://github.com/agno-agi/agno/pull/10666
fix: serialize PEP 604 tool parameter types in UserInputField.todict by @ShaoLOWu in https://github.com/agno-agi/agno/pull/11177
* refactor: rename Product-scoped Company tools by @ashpreetbedi in https://github.com/agno-agi/agno/pull/11242
fix: map Mistral client params onto the mistralai SDK and alias safemode to safe_prompt by @ysolanky in https://github.com/agno-agi/agno/pull/11222
* chore: Release v3.1.9 by @harshsinha03 in https://github.com/agno-agi/agno/pull/11248
New Contributors
* @Fazlul105 made their first contribution in https://github.com/agno-agi/agno/pull/11231
* @ShaoLOWu made their first contribution in https://github.com/agno-agi/agno/pull/11177
Full Changelog: https://github.com/agno-agi/agno/compare/v3.1.8...v3.1.9
v3.1.82026-10-21
Changelog
New Features
- Postgres Sandbox Checkpoint Store: Added PostgresCheckpointStore for storing sandbox checkpoints directly in PostgreSQL, allowing multi-replica deployments to restore sessions without S3 or shared filesystems. Checkpoints are chunked, written transactionally, deduplicated by SHA-256 and verified on read. It is opt-in; existing checkpoint defaults are unchanged. (#11186)
- Framework Managers as Routine Executors: Company Routines can now use the internal Product Manager or OS Manager as executors through declarations, the Routine API and human-chat configuration. Reserved manager IDs resolve only from the owning AgentOS and cannot be replaced by saved or user-defined components. Includes an optional weekly Product Manager review. (#11193)
- Sandbox Checkpoint Retention: Sandboxes now retain the three newest checkpoints per session by default. Configure retain_checkpoints=None to preserve every checkpoint. Unpublished or superseded checkpoints are retired and deleted after the run outcome is safely stored. (#11187)
Improvements
- Sandbox Session Cleanup: Deleting a sandbox-backed session now removes its warm computer and associated checkpoint objects after the database transaction commits. Single, bulk and user-scoped session deletion are supported without affecting sessions the caller does not own. (#11187)
- OS Manager Specialist Visibility: Platform Builder and Platform Manager now appear as members of OS Manager in /teams and /teams/os-manager-internal. They remain excluded from the reusable application registry, top-level agent discovery and standalone agent routes. (#11198)
- Claude Defaults to Sonnet 5.5: Claude() now defaults to claude-sonnet-5-5, replacing the deprecated claude-sonnet-4-5-20250929 before its retirement. Non-default temperature, topp and topk values are dropped with a warning for newer Claude models that reject those sampling parameters. This behavior applies across Anthropic, Bedrock, Foundry and Vertex providers. (#11212)
Behavior Changes
- Checkpoint Retention: Only the three newest sandbox checkpoints are retained by default. Opening a checkpoint for an older run can now raise CheckpointObjectNotFound. Set retain_checkpoints=None to restore the previous unlimited-retention behavior. (#11187)
- Durable Sandbox Startup Validation: AgentOS now fails at startup when a durable-queue sandbox uses the host-local default checkpoint store. Configure a checkpoint store shared by every replica, such as PostgresCheckpointStore or S3CheckpointStore. Previously this invalid configuration failed only on the first submission. (#11187)
Bug Fixes
- Missing Sandbox Checkpoints: A committed checkpoint missing from storage now ends the turn with a clear, non-retryable error, retires the newly created computer and releases the session. Previously the run remained RUNNING, exhausted queue retries and blocked later turns on the session. (#11187)
- Company Routine Execution: Denied status writes and failed Studio runner delegation now correctly fail the Routine. Stale archived catalog entries no longer collide with owned reserved managers. (#11193)
What's Changed
* fix: sandbox checkpoint recovery, retention and session-delete cleanup by @ysolanky in https://github.com/agno-agi/agno/pull/11187
* fix: show platform specialists under OS Manager by @ashpreetbedi in https://github.com/agno-agi/agno/pull/11198
* feat: add PostgresCheckpointStore for sandbox checkpoints in Postgres by @ysolanky in https://github.com/agno-agi/agno/pull/11186
* feat: support framework managers as Company Routine executors by @ashpreetbedi in https://github.com/agno-agi/agno/pull/11193
* fix: default Claude to sonnet-5-5 before sonnet-4-5 retires by @kausmeows in https://github.com/agno-agi/agno/pull/11212
* fix: default OpenAITools image generation to gpt-image-2 by @23821 in https://github.com/agno-agi/agno/pull/10824
* chore: split Company db suites out of test-db and add a sixth unit-test shard by @harshsinha03 in https://github.com/agno-agi/agno/pull/11213
* fix: send adaptive thinking when a Claude model rejects a thinking budget by @harshsinha03 in https://github.com/agno-agi/agno/pull/11215
* chore: Release v3.1.8 by @kausmeows in https://github.com/agno-agi/agno/pull/11208
New Contributors
* @23821 made their first contribution in https://github.com/agno-agi/agno/pull/10824
Full Changelog: https://github.com/agno-agi/agno/compare/v3.1.7...v3.1.8
v3.1.72026-10-20
Changelog
Improvements:
- OS Manager Tool Limits: OSManager takes toolcalllimit (default 8, per coordinator run, including delegations) and membertoolcall_limit (default 24, per specialist invocation). Explanations stay scoped to the question and grounded in the Agno docs. Set either limit to None to restore unlimited tool use, as before. See cookbook.
- Cartesia Default Model: CartesiaTools now defaults to sonic-3.6. Cartesia sunsets the sonic-2 snapshots, the previous default, on October 20, 2026.
Bug Fixes:
- Company Routines: Editing a Product's metadata no longer unlinks its Routines. SchedulerTools and Studio can no longer delete, enable, disable or rewrite Company-managed schedules. OS Manager sends Routine and Product work to Product Manager. Under user isolation, Routines with no owner run again, and refresh archives only the caller's finished work.
- Postgres Metrics: calculate_metrics() works past the driver's query parameter limit (32,765 sessions on asyncpg, 65,533 on psycopg). On AsyncPostgresDb, a failed refresh now returns an error instead of a 200 with empty metrics.
- v3 Migration on Postgres: The sessions-to-runs migration streams rows through a server-side cursor. Memory is bounded by one batch, so large sessions tables no longer run out of memory.
- Sandbox Checkpoints with Virtualenvs: A workspace containing a Python virtualenv no longer fails checkpoint capture on its absolute interpreter links. A checkpoint that fails because of the workspace files now ends the run with a non-retryable error and releases the session, which previously stayed stuck in finalizing.
- S3 Checkpoint Cancellation: Cancelling S3CheckpointStore.aopen_archive no longer leaves the downloaded copy in scratch, and the cached-object digest check runs off the event loop.
What's Changed
* fix: Move Cartesia's default TTS model to Sonic 3.6 by @jiangjiang248 in https://github.com/agno-agi/agno/pull/10750
* fix: stream the v3 sessions-to-runs migration on Postgres instead of loading it all into memory by @Himanshu040604 in https://github.com/agno-agi/agno/pull/11183
* fix: checkpoint virtualenv links and release sessions whose files cannot be checkpointed by @ysolanky in https://github.com/agno-agi/agno/pull/11191
* fix: bound OS Manager tool use and ground explanations by @ashpreetbedi in https://github.com/agno-agi/agno/pull/11175
* fix: compute Postgres metrics past the driver query parameter limit by @sannya-singal in https://github.com/agno-agi/agno/pull/11182
* fix: remove S3 checkpoint copies when an async open is cancelled by @ysolanky in https://github.com/agno-agi/agno/pull/11194
* fix: keep Company Routines linked, routed and owned consistently by @ysolanky in https://github.com/agno-agi/agno/pull/11184
* chore: Release v3.1.7 by @kausmeows in https://github.com/agno-agi/agno/pull/11174
Full Changelog: https://github.com/agno-agi/agno/compare/v3.1.6...v3.1.7
v3.1.62026-10-19
Changelog
New Features:
- Eval Case Run Arguments: Case takes sessionid, userid, sessionstate, dependencies, metadata and knowledgefilters, forwarded to the agent or team run, and input accepts a Message, so a case can attach an image or file to the graded turn. Cases sharing a session_id run as consecutive turns. See cookbook.
Improvements:
- OS Manager Docs Access: OS Manager queries the Agno documentation directly for framework questions instead of delegating to Platform Manager, and keeps delegated tasks scoped to the question. access_docs now covers both.
- Result Offloading Search and Readback: searchresult returns separate passages from one long line, such as single-line JSON, instead of only the first match, and readresult takes max_chars to read a smaller excerpt.
Bug Fixes:
- ParallelTools: Copying an agent or team with ParallelTools, as AgentOS does on every request, keeps the live SDK client and gives the copy its own settings, without RLock copy errors or _state finalizer warnings.
What's Changed
* feat: add run arguments and Message input to eval Case by @harshsinha03 in https://github.com/agno-agi/agno/pull/10829
* fix: preserve Parallel SDK clients when copying toolkits by @ashpreetbedi in https://github.com/agno-agi/agno/pull/11172
* fix: keep OS Manager focused on the question by @ashpreetbedi in https://github.com/agno-agi/agno/pull/11169
* fix: retrieve focused passages from offloaded tool results by @ashpreetbedi in https://github.com/agno-agi/agno/pull/11171
* chore: Release v3.1.6 by @kausmeows in https://github.com/agno-agi/agno/pull/11170
Full Changelog: https://github.com/agno-agi/agno/compare/v3.1.5...v3.1.6
v3.1.52026-10-19
Changelog
New Features:
- Repositories for Sandboxed Agents: A sandboxed ClaudeAgent or CodexAgent takes repository=Repository(url, branch="main") and starts every session in that repository at /workspace/repository. The first turn pins the exact commit, later turns keep the pin and the engineer's edits across warm reuse and checkpoint restore, and private repositories authenticate with GITHUBREADTOKEN from the sandbox environment. See cookbook. (#11121)
- S3 Checkpoint Store: S3CheckpointStore(bucket, prefix=..., endpoint_url=...) keeps sandbox checkpoints in an S3-compatible bucket, so sandboxed agents run on several replicas without a shared filesystem. Objects are content-addressed, uploads are verified before put returns, and credentials come from the AWS environment. See cookbook. (#11133)
- Open a Run's Sandbox Checkpoint: openruncheckpoint(runid) / aopenruncheckpoint on a sandboxed agent yield the verified archive a completed run committed, so a host process can read what a turn left behind after the container is gone. checkpointstore exposes the store the agent's turns write to. See cookbook. (#11131)
- On-Demand Schedules: A schedule with an empty or omitted cronexpr never arms itself and runs only through its trigger route, over HTTP and through ScheduleManager. ScheduleResponse.cronexpr is now optional and reports it as null. (#11118)
- KeenableTools: New toolkit for Keenable web search and page extraction (websearch, fetchurl_content, with async variants). It works with no API key and no new dependency, and refuses fetches that resolve to private or loopback addresses. See cookbook. (#9558)
Improvements:
- External Agents in the Registry: Registry(agents=[...]) accepts ClaudeAgent and CodexAgent, and AgentOS agent, team and workflow responses carry a sandboxed flag. (#11127)
- Schedule Endpoint Edits: Changing a schedule's endpoint over PATCH /schedules/{schedule_id} or through ScheduleManager now updates its target columns instead of leaving them stale. (#11118)
Bug Fixes:
- AgentOS Form Defaults on FastAPI 0.147: FastAPI 0.147.0 reads omitted bool form fields as False, so run and continue routes returned JSON instead of SSE when stream was omitted and continues without regenerate=true failed with 400. The routes now restore the declared defaults for stream, monitor and replace_original. (#11160)
- Sandbox Runs Without Redis: Sandboxed agent runs over POST /agents/{agent_id}/runs no longer answer 400 on a Postgres durable queue without Redis. Cross-replica streaming and cancellation still need Redis. (#11161)
- Workflow Files: File outputs produced inside Parallel, Steps, Loop, Condition and Router steps now reach the next step and workflowrunresponse.files. (#10950)
- YouTubeTools: Video ids are read from Shorts, live, mobile, music and privacy-enhanced URLs, and empty ids return No video ID found before any request. (#10707)
- PerplexitySearch and PubmedTools: An explicit max_results=0 is honoured instead of being replaced by the default. (#10595)
- InMemoryDb: upsertusermemory stores a copy, so mutating the UserMemory after the upsert no longer changes the stored memory. (#11016)
What's Changed
* feat: add Company heartbeat routines and sources by @ashpreetbedi in https://github.com/agno-agi/agno/pull/11102
* feat: automate Ready Tasks and bounded Routine acceptance by @ashpreetbedi in https://github.com/agno-agi/agno/pull/11104
* feat: configure repositories for sandboxed harness agents by @ashpreetbedi in https://github.com/agno-agi/agno/pull/11121
* fix: complete Company automation and configuration follow-up by @ashpreetbedi in https://github.com/agno-agi/agno/pull/11118
* fix: stabilize Company Task session ownership and recovery by @ashpreetbedi in https://github.com/agno-agi/agno/pull/11126
* feat: execute Company Tasks with sandboxed harness agents by @ashpreetbedi in https://github.com/agno-agi/agno/pull/11127
* [fix] propagate files through workflow container steps by @Timotty in https://github.com/agno-agi/agno/pull/10950
* [feat] Add KeenableTools for keyless web search and page extraction by @ilya-bogin-keenable in https://github.com/agno-agi/agno/pull/9558
* fix: read the video ID from Shorts, live and mobile YouTube URLs by @devYRPauli in https://github.com/agno-agi/agno/pull/10707
[fix] Honor explicit maxresults=0 in PerplexitySearch and PubmedTools by @ryanakml in https://github.com/agno-agi/agno/pull/10595
* fix: detach the stored user memory from the caller's object by @insuffer in https://github.com/agno-agi/agno/pull/11016
* feat: open a run's committed sandbox checkpoint through the agent by @ysolanky in https://github.com/agno-agi/agno/pull/11131
* feat: add S3CheckpointStore for sandbox checkpoints in object storage by @kausmeows in https://github.com/agno-agi/agno/pull/11133
* fix: allow sandbox API runs on a durable queue without Redis by @ashpreetbedi in https://github.com/agno-agi/agno/pull/11161
* fix: restore AgentOS boolean form defaults under FastAPI 0.147 by @harshsinha03 in https://github.com/agno-agi/agno/pull/11160
* feat: create Company Products and Routines from the OS Manager chat by @ashpreetbedi in https://github.com/agno-agi/agno/pull/11159
* chore: Release v3.1.5 by @kausmeows in https://github.com/agno-agi/agno/pull/11155
New Contributors
* @Timotty made their first contribution in https://github.com/agno-agi/agno/pull/10950
* @ilya-bogin-keenable made their first contribution in https://github.com/agno-agi/agno/pull/9558
* @ryanakml made their first contribution in https://github.com/agno-agi/agno/pull/10595
* @insuffer made their first contribution in https://github.com/agno-agi/agno/pull/11016
Full Changelog: https://github.com/agno-agi/agno/compare/v3.1.4...v3.1.5
v3.1.42026-10-16
Changelog
New Features:
- Sandboxed Harness Agents: ClaudeAgent and CodexAgent take a sandbox= parameter and run the whole harness process inside an isolated, per-session computer instead of on the host, with session state that survives losing the container, the host or the replica that ran the last turn. No new classes to hold: ClaudeAgent(id="coder", db=PostgresDb(...), sandbox=DockerSandbox("agno-sandbox:local", network="isolated")) is the whole setup, and the agent builds its coordinator from its PostgresDb on first use. Each turn claims the session through a lease row in the agent's Postgres (<sessiontable>sandboxbindings, created on first use), starts compute restored from the session's last committed checkpoint (workspace plus the harness's own conversation files), runs the turn, then captures and verifies a new checkpoint and publishes the run row and recovery point in one transaction. A cancelled or failed turn publishes nothing and the previous checkpoint stays the recovery point. The turn refuses to continue on state it cannot verify: it passes the native conversation id (Claude session id, Codex thread id) committed with the last checkpoint, and a restore that lost it fails with nativestatemissing before the model runs instead of silently starting a new conversation. Typed harness settings travel to the runtime as JSON; host-only settings (cwd, retries, env, native option objects) are rejected by name at construction. requirenativeresume and strictpersistence are new options on BaseExternalAgent, and agno[claude] / agno[codex] install extras cover the harness SDKs. See cookbook. (#11028, #11043, #11056, #11057)
- Docker Sandbox: DockerSandbox is the customer-operated provider. Containers run as UID 1000 with a read-only image filesystem, dropped capabilities, no privilege escalation, CPU, memory and process limits, bounded tmpfs for /workspace and /runtime, no host or Docker socket mounts, and only explicitly configured environment values plus the harness's model API key. Networking defaults to none; network="isolated" gives a managed bridge with inter-container traffic off, and subnet= picks its range. Deterministic session and generation identity reconciles creation races and lost responses; commands are never retried automatically. pause / resume are provided through the kernel freezer, and DockerCheckpointTransfer moves checkpoints in and out. cookbook/frameworks/sandbox/build_image.py builds the runtime image. See cookbook. (#10975, #11028)
- Amazon Bedrock AgentCore Sandbox: AgentCoreSandbox implements the same contract on AgentCore Runtime, with AgentCoreCheckpointTransfer next to the Docker transfer. Each sandbox generation is one AgentCore runtime session in a Firecracker microVM; create is idempotent through a session id derived from the namespace, agent, session and generation. A root server on port 8080 claims the microVM, starts the runtime as UID 1000, fences every later operation with the claimed configuration and a random incarnation so a stale handle can never act on a later claim, forwards start and poll calls, and runs commands behind a guard that drops to UID 1000 in /workspace. Checkpoints stop UID 1000 processes during capture and restore and move through InvokeAgentRuntime in chunks, up to 2 GiB. Pause is not supported and AgentCore session storage is not used. The provider packages moved to agno.sandbox.docker and agno.sandbox.agentcore; DockerCheckpointTransfer now imports from agno.sandbox.docker. The runtime also gained two Codex fixes that apply to Docker as well: it creates CODEXHOME, and it logs Codex in from OPENAIAPIKEY when no stored login matches, since checkpoints exclude auth.json. cookbook/frameworks/sandbox/agentcore/agentcoresetup.py builds, pushes and deploys the runtime. See cookbook. (#11050)
- Warm Sandbox Turns: The coordinator keeps a session's computer for warmcomputerseconds (default 300) after a turn instead of destroying it: paused on Docker, left idle on AgentCore. The next turn inside that window resumes it and restores nothing; otherwise it retires the old computer and takes the cold path on a fresh generation restored from the checkpoint. The checkpoint is still captured after every turn, so a warm computer is only a shortcut to the state the checkpoint already holds. metadata.sandbox.resumed on every completed run says which path the turn took, and retirecomputer / sweepwarm on the coordinator remove computers that sessions never came back for. Live on Docker a warm Claude turn took 4.6 s against 8.2 s cold; on AgentCore 3.3 s against 8.2 s. See cookbook. (#11088)
- Sandbox Through AgentOS: POST /agents/{agentid}/runs for a sandboxed agent submits foreground, streaming and background requests to the durable queue with no new endpoints. An Idempotency-Key retry returns the same run, while a different session, input, options or attachment bytes returns 409. Clients reconnect with /runs/{runid}/resume (lasteventindex) and cancel with /runs/{run_id}/cancel, including across replicas: a cancel settled on another replica now records the run as cancelled and keeps the user's turn on the run row. AgentOS fails at startup if a sandboxed agent has no Postgres to coordinate on, refuses the host-local default checkpoint store once a queue worker is active, and validates each sandboxed agent's bindings table at startup and before a submission, naming the remedy for a table from an earlier build. See cookbook. (#11043, #11105)
- Attachments for Sandboxed Agents: run / arun on a sandboxed ClaudeAgent or CodexAgent accept images= and files=, and the AgentOS run route accepts them as uploads. Only bytes cross into the sandbox: each attachment is validated and encoded before any durable write, staged under /workspace/.agno/uploads/<run_id>/ by the harness's own staging, and kept in the turn's checkpoint so later turns on a new computer can still read it. A request carries at most 16 attachments and 512 KiB in total; URLs, media references, server file paths, audio and video are refused. Queued runs store the attachments with the submission, so a retry sends the same bytes. See cookbook. (#10990, #11091)
- Images and PDFs Inline for Claude Code: ClaudeAgent now sends attached images (PNG, JPEG, GIF, WEBP) and PDFs up to 4 MiB inside the message as content blocks, the form the Agent SDK's streaming input takes, instead of only writing them to disk for Claude's Read tool. Claude sees the picture directly and never reads the file back, which used to return the whole file as base64 and overflow the SDK's 1 MiB message buffer and the sandbox's 256 KiB event frame for images of a few hundred KB. Whether a file goes inline is decided from its bytes, so a mislabelled file stays on disk for the file tools; other files and larger images keep the Read path. See cookbook. (#11115)
- Queue Per Session: Durable background runs on one session now execute at most one at a time, in submission order, for every component. Each ticket carries a store-assigned session_seq; a ticket is claimed only while no sibling in its session is running or