Changelog

Features

• feat(library): GitHub MCP Server for AI Agents: Builders, Security, and Setup
• feat(library): Best Slack MCP Agent Builders: 7 Options Compared
• feat(library): How to Use the GitHub MCP Server With AI Agents: Builders, Security, and Setup

Bug Fixes

• fix(db): compare tables by schema-qualified name in the migration audit
• fix(deps): patch sharp librsvg CVE and MCP SDK OAuth issuer binding
• fix(db): refuse migrations whose ownership statements only pass outside production
• fix(build): keep zod's default error messages in production builds

Other Changes

• docs(library): update govern-ai-agents-multiple-teams-enterprise-workspace
• docs(library): update best-conversational-ai-platforms-crm-helpdesk
• docs(library): update what-is-human-in-the-loop-in-ai-agents
• ci: bump the actions group across 4 directories with 19 updates
• docs(library): update open-source-ai-agent-platforms

View changes on GitHub

Features

• feat(buffer): complete stable API tool coverage
• feat(api): expose credential sharing and SSO administration
• feat(models): add current frontier models
• feat(integrations): add Ramp and Microsoft Intune
• feat(permission-groups): add agent default
• feat(providers): add Nebius Token Factory
• feat(desktop): import local files from the background executor
• feat(desktop): keep the machine awake for background chats and yield the page to the user
• feat(desktop): show background desktop chats and keep the chat view display-only for them
• feat(desktop): run a chat's desktop tools in the background executor
• feat(mothership): let the worker choose models when the model picker is off
• feat(desktop): bind turns to a desktop and enforce its deadlines from the row
• feat(sso): allow self-hosted DNS verification bypass
• feat(plane): add core integration with self-hosted support
• feat(library): 25 Multi-Step AI Agent Examples Beyond Chatbots
• feat(desktop): device registry, inbox and leased claims for a background executor
• feat(library): Best Conversational AI Platforms With CRM and Helpdesk Integrations
• feat(library): Best AI Coding Agents for Refactoring and Debugging: 5 Tools Compared
• feat(mothership): keep each chat's reasoning effort, default to medium, restore Low
• feat(projects): add project identity and lifecycle foundation
• feat(workflows): stop a manual v2 run after a block, and workflows run --stop-after
• feat(library): How do you build an AI bot for Discord without code?

Improvements

• improvement(desktop): ship the background executor without a flag, close its QA gaps
• improvement(files): add sharing to file resource view
• perf(pii): batch spaCy NER passes in redact_batch
• improvement(ci): check out same-repo pull requests through a git mirror
• improvement(mothership): relabel the Sol pick GPT-6.1 Sol and retire the none effort
• refactor(mothership): a saved Stop handoff's id wins on restore
• refactor(mothership): keep a queued send's reused id in one field
• refactor(mothership): one hold field and one retry field on a queued send
• improvement(desktop): Electron E2E for desktop tool lifetimes against a live local Sim
• refactor(mothership): one re-queue policy and one send payload, and keep a Stop-failed send waiting for the user
• refactor(desktop): one forward-only state per recorded tmux run, and one check for a delivered result
• improvement(outbox): load handler modules only for the event types a run will process
• improvement(cron): only start outbox and file-search passes when work is due, via a shared scheduled-pass helper
• improvement(triggers): move TRIGGER_TYPES to a dependency-free module
• perf(cli): load a manual run's draft once and keep embedded CLI results lean
• improvement(trigger): downsize connector sync and document processing machines
• improvement(tables): read row counts and versions through an append-only change log
• perf(sandbox): grant the run_code session lease in the reconnect instead of extra E2B round trips

Bug Fixes

• fix(code-placeholders): reject arithmetic comparisons and heredoc operands
• fix(desktop): poll desktop activity only for users with a desktop, keep XDG zsh configs integrated
• fix(audit): harden data drains and expand security event coverage
• fix(executor): preserve stop-after across pause and resume
• fix(billing): converge Stripe subscription syncs and stop webhook echoes overwriting unsynced changes
• fix(docs): publish full LLM text as a static asset
• fix(docs): prerender complete LLM documentation
• fix(tables): stop the fold sweep from starting a page query past its budget
• fix(mothership): cancel desktop tools a signed-out turn starts late, guard the shown-once license key
• fix(sandbox): keep reused E2B workbench leases within the runtime cap
• fix(settings): protect generated keys and identity transitions
• fix(mothership): use a spinner while workflow resources load
• fix(mothership): close a pre-aborted SSE stream, keep the new-chat effort across a failed first send
• fix(desktop): stop a run the model never got the result of
• fix(outbox): log unloaded handler modules as a failure, not a completed run
• fix(desktop): linear path trim, prompt Stop between keystrokes, clear a closed session's input marker
• fix(ci): count both sides of a rename in the desktop live gate, keep canary reports
• fix(desktop): keep a chat-view import alive while it works, by the lease its session renews
• fix(mothership): a pure resend verdict, and Send-now drops a message the server already has
• fix(mothership): send a late queue write to the chat a new-chat queue moved to
• fix(desktop): leave a stopped native file tool's outcome to Stop
• fix(mothership): keep a failed direct send ahead of follow-ups queued during its POST
• fix(mothership): decide once whether a queued send may already be on the server
• fix(desktop): frame each tmux format record so a newline in a field cannot forge a pane
• fix(desktop): stop the agent's tmux runs a previous process left going
• fix(mothership): retry a send that failed while the browser stayed online
• fix(mothership): claim under the chat lock in the Stop-versus-recovery settlement test
• fix(ci): tidy stop-session.sh scans, listings and arguments
• fix(settings): prevent false unsaved prompts and preserve drafts
• fix(mothership): keep any queued send the server may already hold from being edited
• fix(desktop): run an agent command on tmux that cannot tag its pane, untracked
• fix(ci): match E2E app tags literally, make them unique per run, and never stop the caller
• fix(mothership): keep a withdrawn first message held at the queue head as written
• fix(events): authorize the events queued before a stream opens once
• fix(desktop): read tmux format output with a separator no tmux version escapes
• fix(mothership): run same-workflow workflow tools in order instead of failing as busy
• fix(webhooks): skip duplicate webhook deliveries already in progress instead of polling
• fix(ci): stop the detached telemetry flush too, re-signal stragglers on a monotonic clock, and report HTTP E2E failures by route
• fix(mothership): keep a chatless surface's queue across remounts and contain hung chat hook tests
• fix(events): start every SSE response once its subscriptions are live
• fix(chat): resolve a new chat's history entry to the chat route on Back, and say when an effort save fails
• fix(analytics): protect production Google measurement
• fix(mothership): close the remaining ways a chat send is lost, resent hot, or polled after unmount
• fix(e2e): compile every timed route first in the HTTP suites and stop sessions completely
• fix(mothership): prune composer contexts during render so fast typing cannot trip the update-depth limit
• fix(plane): validate webhook fields by registration mode
• fix(ci): stop each HTTP end-to-end app fully and start the next from an empty dev cache
• fix(plane): correct tile styling and capped connector syncs
• fix(mothership): keep a message the server never admitted instead of dropping it
• fix(mothership): re-attach at once when the user returns to a recovered stream
• fix(mothership): re-read the transcript until the server has saved a finished turn
• fix(desktop): inbox persistence order, desktop-only overdue sweep, and ungated pickup windows
• fix(mothership): end desktop tools at sign-out and settle held reports as final
• fix(copilot): store tool file outputs from workspace chats under the Copilot user
• fix(mothership): keep local file tools running when the chat view changes
• fix(mothership): keep a new chat's effort pick changed while its first send is pending
• fix(sandbox): judge session lease coverage from the request's own clock reading
• fix(mothership): fail unclaimed desktop calls fast and stop dropping them silently
• fix(mothership): default chat effort back to high
• fix(projects): revoke banned users' keys first and create imported workflows atomically
• fix(executor): unwrap boxed primitives by internal slot when checking JSON serializability
• fix(mcp): treat different URL credentials as a new destination
• fix(mothership): roll back a failed effort save by pick, not by value
• fix(mothership): refuse desktop claims for unapproved or stopped calls
• fix(seo): make content pagination indexable
• fix(projects): restore workspace deletion and tighten Project lifecycle
• fix(executor): fail a stop-after run whose routing skips the stop block
• fix(credentials): keep the field-less GitHub App installation out of v2 provider discovery
• fix(code-placeholders): reject shell arithmetic placeholders
• fix(mcp): restrict MCP server destination changes to admins
• fix(logs): keep a block log's file size from changing after the block completes
• fix(executor): drop the full JSON clone of execution state at run completion
• fix(chat): label simple effort options with the effort they send

Other Changes

• ci: duration-balanced integration shards, warm desktop-live, CI on every PR
• ci: faster, cheaper, consistently named CI
• test(desktop): pin the journal snapshot before recovery, keep start going when the journal cannot load, and run tmux in the canary E2E
• test(desktop): stop the agent's terminal commands in the real app, with real shells and real tmux
• test(mothership): stop the online case of the accepted-resend test passing without its history check
• chore(nextjs): upgrade to 16.4.0
• chore(trigger): fix queues that never serialized and schedule missing self-hosted crons
• chore(table): remove unused table-update background task
• docs(library): update 6-best-ai-observability-tools-for-production-agents-in-2026
• test(desktop): keep the SQL and TypeScript desktop-call classifiers in agreement
• docs(library): update best-ai-agents-for-lead-enrichment-2026
• docs(library): update sim-vs-dedicated-chatbot-builders
• docs(library): update top-ai-assistants-2026
• docs(library): update ai-agent-examples-by-department-and-industry
• docs(library): update open-source-ai-agent-platforms
• docs(library): update best-multi-agent-frameworks-2026
• test(mcp): exercise the real destination check in lifecycle tests
• docs(workflows): describe what still runs when a stop-after target is ruled out
• docs(library): update sim-open-source-zapier-alternative
• docs(library): update ai-agents-in-procurement
• docs(library): update mcp-security
• docs(library): update aeo-vs-geo-what-answer-engine-and-generative-engine-optimization-actually-mean
• docs(library): update ai-agent-observability
• docs(library): update best-ai-agents-for-customer-support-automation
• docs(library): update dify-alternatives
• docs(library): update langgraph-alternatives
• chore(terms): update terms of service
• docs(library): update agentic-ai-coding-tools-what-they-are-and-how-the-top-options-compare

View changes on GitHub

A Helm chart for Sim - the open-source AI workspace where teams build, deploy, and manage AI agents

Bug Fixes

• fix(mothership): preserve workflow identity in background confirmations
• fix(files): retain known lineage through binary exports and extraction
• fix(code-placeholders): make python bare-placeholder classification and tagged-template edits linear
• fix(execution): make code placeholder and reference scans linear

View changes on GitHub

Features

• feat(library): Slack AI Agent Security: Permissions, Data, and Audit Guide
• feat(library): Best HubSpot Alternatives for AI Marketing Automation in 2026
• feat(library): How do you build an AI voice agent with Twilio and Sim?

Improvements

• improvement(knip): check entry exports of private packages and delete their dead barrels
• improvement(tools): stop exporting tool types used only in their own file
• improvement(tools): delete unreferenced tool response types and the dead generic resource data path
• improvement(guidance): make every ratchet shrink-only, scope rules to their files, fix stale guidance

Bug Fixes

• fix(tools): reject dot path segments in tool request URLs
• fix(webhooks): check existing path owners before claiming on deploy
• fix(code-placeholders): skip heredoc bodies when scanning shell quote context
• fix(tools): check usage limits before running hosted-key tools via the API
• fix(docs): cap search query length and time out the embedding call

View changes on GitHub

Features

• feat(forks): compare last synced source deployments
• feat(youcom): add You.com integration
• feat(search): browse Lucid folders and Notion pages
• feat(library): How to Use a Postgres MCP Server With AI Agents: Security, Deployment, and Evaluation
• feat(powerbi): add Power BI actions

Bug Fixes

• fix(docs): document You.com tools and expand shared output spreads
• fix(credentials): make Claude Platform API keys available in the provider catalog
• fix(audits): apply browser-runtime rules to @sim/utils; align settings checklist with the standalone description rule
• fix(files): read offset 0 as the first line instead of rejecting it
• fix(landing): return 404 for unknown library, blog, integration, model, and author slugs
• fix(mothership): make chat fork complete, consistent and safe to retry
• fix(sdk): default both SDKs to the www.sim.ai host
• fix(integrations): list tools for integrations without an operation dropdown
• fix(knowledge): scope list counts to selected knowledge bases
• fix(audits): close guardrail detector gaps and correct guidance from release review
• fix(powerbi): clear the explicit-any and unused-export ratchets on staging
• fix(mothership): leave desktop tool calls to the desktop app in other clients

Other Changes

• chore(pi): upgrade agent to 1.0.0
• docs(library): update ai-agent-examples-by-department-and-industry

View changes on GitHub

TypeScript SDK v0.2.1

Published simstudio-ts-sdk@0.2.1 to npm.

Installation

npm install simstudio-ts-sdk@0.2.1

Documentation

See the README or the docs for more information.

Python SDK v0.2.1

Published simstudio-sdk==0.2.1 to PyPI.

Installation

pip install simstudio-sdk==0.2.1

Documentation

See the README or the docs for more information.

Improvements

• improvement(audits): enforce console, helper, render-path, persist, and deployment-flag rules

Bug Fixes

• fix(deps): bump mammoth to 1.12.3
• fix(logging): name the driver cause and redact bound params in logged errors
• fix(files): resolve chat uploads whose names are not in VFS form
• fix(search): reuse managed MCP sessions within operations

Other Changes

• docs(agents): accurate guidance, guardrail map, and check:guidance-refs

View changes on GitHub

Features

• feat(workflows): compare deployment versions across UI and API
• feat(files): render diff fences and theme mermaid diagrams
• feat(dashboards): embed live dashboard panels in markdown
• feat(chat): cycle available composer modes with Shift+Tab

Improvements

• improvement(audits): ratchet unused exports, explicit any, and file names
• improvement(lint): strip comment noise and add check:comment-hygiene
• improvement(cron): run stale execution and file version cleanup off the request path

Bug Fixes

• fix(db): pin the primary-key plan in the search retirement read-bound test
• fix(cli): report authentication accurately and preserve file text
• fix(files): align workbench reads with provenance policy
• fix(accounts): report managed OAuth failures accurately
• fix(slack): honor implicit Search approval during authorization
• fix(mothership): bound stream recovery with a per-run restart budget
• fix(slack): verify Search permissions before changing active grants
• fix(search): correct onboarding and live citations
• fix(dashboards): keep DST fall-back range endpoints and resolve threshold axis from first series
• fix(ci): isolate HTTP end-to-end suites and stop the SCIM readiness flake
• fix(library): ignore sentence punctuation after bare internal URLs in check:library-content
• fix(mothership): stop duplicating chat resource tabs in workspace chats
• fix(sandbox): preserve workbench availability for unrecorded API responses
• fix(chat): preserve per-chat resource panel widths
• fix(wiza): surface Wiza's nested error messages
• fix(accounts): scope organization OAuth outbound requests
• fix(library): scope Sim license claims to the Apache 2.0 core and the Sim Enterprise License
• fix(seo): escape backslashes in llms-full.txt comparison table cells

Other Changes

• chore(lint): delete unused variables and parameters; enforce both rules repo-wide
• chore(cleanup): apply codebase-design deep-module pass across the monorepo
• chore(search): add paced projection replacement and data retirement
• chore(search): remove legacy indexed enterprise search
• chore(content): add check:library-content audit for content posts
• docs(library): restore content dropped by automated refreshes

View changes on GitHub