Features

• feat(billing): require a paid plan or prepaid credits for Chat (#9551)

Bug Fixes

• fix(billing): fail closed on legacy Chat payer reads and cover the Chat plan gate (#9553)
• fix(public-shares): stream shared files with byte ranges instead of buffering them under 100 MB (#9550)
• fix(sandbox): initialize workbench history before honoring a stop during creation (#9549)

Contributors

• @waleedlatif1

View changes on GitHub

Features

• feat(library): AI Agent Output Controls Compared: Guardrails, Structured Outputs, and Streaming (#9539)
• feat(library): Best AI Agents for Jira Triage and Datadog Incident Response (#9536)
• feat(mothership): answer a long workflow tool run still running, and stop detached runs on Stop (#9507)
• feat(people): organization People page with profiles, reporting lines, and people-data import (#9517)
• feat(desktop): continue a passkey sign-in in your browser and bring the session back (#9515)
• feat(mothership): erase private memory when a member leaves or an organization is deleted (#9503)

Improvements

• improvement(docs): collect reader feedback inline (#9538)
• improvement(release): harden favicons, sim links, and the extension from a pre-release audit (#9530)
• improvement(search): overlap live search stages and report every unsearched source (#9514)
• improvement(desktop): sign-in prompts for the agent browser — site passwords, passkey accounts, client certificates, and app links (#9521)
• improvement(cli): sim credentials providers lists providers (#9522)
• improvement(desktop): install updates on one click, asking only about real in-flight work (#9505)
• improvement(cli): files batch-delete takes file ids as arguments (#9518)
• improvement(desktop): bring agent browser menus, keys and fullscreen in line with Chrome (#9512)
• improvement(extension): document browser data handling and limited use
• improvement(extension): document browser data handling and limited use (#9513)
• improvement(oauth): serve still-valid tokens while refreshing ahead of expiry (#9509)
• improvement(sim-links): pass queries through, trim pasted punctuation, and suggest links from the address bar (#9508)

Bug Fixes

• fix(files): share one served-bytes predicate between the serve route and large downloads (#9548)
• fix(release): read Excel start dates, refuse impossible times and repeated columns, keep typed escapes, no suggestions for punctuation (#9545)
• fix(oauth): default a leader-lock follower's wait to the lease so its takeover can land (#9546)
• fix(desktop): address release review — count app downloads and unchecked terminals before an update restart, send the page referrer on Save Image, and tighten sign-in and people edge cases (#9547)
• fix(ci): drop the duplicate --accept-routes the Tailscale action already passes (#9543)
• fix(files): stream served files with byte ranges instead of buffering them under 100 MB (#9542)
• fix(files): resolve v2 file paths only on the v2 surface, through the canonical resolver (#9531)
• fix(oauth): take over a lapsed refresh lease, share overlapping service-account key reads, and label unsearched accounts (#9537)
• fix(auth): check a workspace's personal-key switch after membership (#9540)
• fix(oauth): keep a five-minute floor on served tokens and refresh ahead earlier (#9528)
• fix(mothership): Stop only cancels detached runs still running, and a held workflow run never blocks the next turn's workbench (#9532)
• fix(desktop): ask before an update restart from the menu, count uploads and downloads as running work, and harden sign-in prompts (#9533)
• fix(files): authorize the workspace before resolving a files/ path (#9527)
• fix(auth): block only curated disposable domains, and never gate SSO or SCIM provisioning (#9529)
• fix(people): keep a suspended manager link on edit and never fail the org layout on the People lookup (#9526)
• fix(execution): name the signal exit code of a failed sandbox process (#9523)
• fix(extension): keep site access compatible with strict CSP (#9524)
• fix(mothership): say why a workflow copilot chat cannot read other chats (#9520)
• fix(auth): refuse account creation from disposable email domains for every provider (#9516)
• fix(desktop): keep sign-in popups working in the agent browser (#9506)
• fix(files): accept files/ VFS paths in v2 file describe, rename and delete (#9504)

Other Changes

• chore(ci): reach the database over the tailnet when CI_DB_TAILNET is set (#9541)
• chore(search): retire legacy search knowledge base projection rows in place (#9511)
• chore(auth): correct where the short disposable list runs (#9535)
• docs(library): update best-multi-agent-frameworks-2026 (#9534)
• Merge pull request #9519 from simstudioai/codex/extension-privacy-production

Contributors

• @waleedlatif1

View changes on GitHub

Features

• feat(favicons): one server route for every site icon, with parent-domain fallback (#9494)

Improvements

• improvement(favicons): resolve platform tenants no provider can see, retry timeout-derived icons, no miss flash (#9497)
• improvement(docs): simplify workflow lightbox and retire migration artifacts (#9496)

Bug Fixes

• fix(docs): restore container expansion and verify canvas interactions (#9502)
• fix(favicons): let a mounted icon retry once its failure ages out (#9501)
• fix(knowledge): keep a bundle's document token totals on import (#9495)

Contributors

• @waleedlatif1

View changes on GitHub

Features

• feat(access-control): govern access requests per permission item (#9483)
• feat(docs): rebuild documentation with the shared Sim design system (#9459)
• feat(extension): open typed sim/ links through the Sim link resolver (#9486)
• feat(sim-links): organization Sim links (go links) with sim/ redirects (#9484)
• feat(library): Best AI Agents for Automated Data Reports in 2026: 4 Tools Compared (#9472)
• feat(library): Best AI Agent Platforms for Slack and Discord: 4 Cross-Channel Options Compared (#9473)
• feat(library): Best AI Agents for Internal Knowledge Search Across Docs and Tools (#9471)
• feat(extension): bring Sim chat and page context to Chrome (#9454)
• feat(mcp): type mcp block replies from declared output schemas (#9466)
• feat(workflows): let a synchronous run continue in the background after a wait timeout (#9464)
• feat(library): Best AI Agent Builders for Automated Daily Email Digests (#9442)
• feat(library): Best AI Agents for Meeting Notes and Follow-Ups: 6 Options Compared (#9443)
• feat(search): unify organization search and reviewed assistant actions (#9428)
• feat(files): page, search, look up and roll up workspace files over the session API (#9435)
• feat(db): seal account tokens by default and backfill them during migrate (#9425)

Improvements

• improvement(sim-links): cleanup and simplify pass on the resource layout (#9492)
• improvement(sim-links): use the shared resource table layout (#9491)
• improvement(sim-links): show each link's destination favicon (#9489)
• improvement(auth): unify sign-in and access pages (#9461)
• improvement(comparisons): lead the OpenAI AgentKit page with the Agent Builder shutdown migration (#9460)
• improvement(sandbox): pause idle chat workbenches instead of keeping them running (#9452)
• perf(knowledge): reconcile a completed listing in one pass over its windows (#9449)
• perf(admin): read dashboard ledger usage one bounded scan per entity (#9448)
• perf(db): serve workspace usage drill-downs index-only on usage_log (#9447)
• perf(audit-logs): read a listing narrowed only by time through per-workspace keyset scans (#9450)
• perf(outbox): resume each retention batch where the previous run's batch ended (#9446)
• perf(file-search): probe each backfill page's revisions by primary key (#9445)
• perf(audit-logs): index organization-level rows and read them as two keyset parts (#9437)
• perf(files): page the Files browser and resolve files without listing the workspace (#9436)
• perf(db): index a workspace's live files in upload order (#9434)
• perf(db): cut hot-path database work and close copilot cleanup gaps (#9419)

Bug Fixes

• fix(docs): simplify selection and verify media interactions (#9493)
• fix(search): accept Fireflies data-array responses (#9490)
• fix(integrations): align connection header actions (#9487)
• fix(search): trace live provider latency and partial failures (#9482)
• fix(mcp): honor json schema openness in mcp output shapes (#9476)
• fix(search): harden assistant review and account discovery (#9478)
• fix(sandbox): correct the E2B build default in the Pi sizing doc (#9477)
• fix(extension): isolate embedded pages from tracking (#9480)
• fix(mothership): settle a sim_cli wait that outlasted --wait-timeout as done, not failed (#9469)
• fix(search): remove the extra assistant tools gate (#9468)
• fix(executor): bound condition loops and reject unnamed variable assignments (#9463)
• fix(search): unify navigation and preserve action review history (#9467)
• fix(billing): hold free plans to their allowance mid-run (#9462)
• fix(integrations): one integration-access verdict across the catalog, @ menus and search, enforced by an audit (#9457)
• fix(search): inline action review and unify account menus (#9451)
• fix(workflows): refuse new placements of sunset blocks in workflow edits (#9456)
• fix(billing): correct the Pi sizing note on E2B's resource maximums (#9455)
• fix(billing): price Pi sandboxes at their build resources (#9453)
• fix(audit-logs): list an organization's audit log through its single scope predicate again (#9438)
• fix(cleanup): keep shared chat files, sweep inline images by prefix, and make room refetches wait (#9433)
• fix(scim): harden directory lifecycle and clarify group mappings (#9429)
• fix(billing): serialize Gmail inbox allowance decisions so racing alias signups can't both keep it (#9431)
• fix(ci): fail the promote wait on a digest lookup error instead of retrying it (#9430)
• fix(files): cap how long a room delivery waits for its refetches (#9427)
• fix(library): merge duplicate sections and FAQs in the ecommerce and Zapier alternatives posts (#9424)
• fix(desktop): hold the GitHub credential check open while the e2e asserts its pending state (#9420)
• fix(billing): withhold the free allowance from Gmail aliases of an existing account (#9418)
• fix(ci): predict the promoted app digest instead of trusting a read right after the retag (#9416)

Other Changes

• chore(search): diagnose unsupported Fireflies response shapes (#9488)
• chore(trigger): run tasks on the bun-1.4 runtime (#9481)
• docs(billing): scope the usage_log workspace index note to created_at windows (#9479)
• test(mothership): drive a draft run past a wait slice through the real controller (#9475)
• docs(library): update 6-best-ai-observability-tools-for-production-agents-in-2026 (#9470)
• chore(deps): bump re2js to 2.8.7 (#9444)
• docs(library): update ai-agent-vs-chatbot (#9441)
• Merge pull request #9426 from simstudioai/staging
• chore(deps): drop the stray bun tag hunk from the zod patch (#9432)
• chore(ci): authenticate Trigger.dev deploys with per-environment API keys (#9423)
• chore(deps): replace deprecated OpenTelemetry trace SDK packages with @opentelemetry/sdk-trace (#9421)
• chore(deps): dependency hygiene cleanup (#9422)
• docs(cli): add agent skills setup to the CLI overview (#9417)

Contributors

• @emir-karabeg
• @waleedlatif1

View changes on GitHub

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

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

Features

• feat(library): Best AI Agents for Drafting Social Media Posts: 4 Platforms Compared (#9414)
• feat(library): Best AI Agent Platforms for Customer Onboarding: A Buyer’s Guide (#9412)
• feat(library): 7 Best AI Assistants for Inbox Management and Reply Drafting in 2026 (#9411)

Bug Fixes

• fix(ci): bound a stalled bun install and let the helm test probe wait for routing (#9409)

Other Changes

• chore(tls): drop the CA distrust step from the Bun services now that Bun ships the blocklist (#9413)
• docs(library): update best-ai-agents-ecommerce-operations-2026 (#9410)
• chore(db): drop the job log schedule index the occurrence unique index covers (#9398)

Contributors

• @waleedlatif1

View changes on GitHub

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

Features

• feat(db): add manual Trigger.dev task for the account token sealing backfill (#9393)
• feat(tailscale): expand and validate API operations (#9385)

Improvements

• improvement(changelog): lead with knowledge base portability (#9377)

Bug Fixes

• fix(ci): only migrate an environment from its own branch (#9395)
• fix(ci): queue migrate runs per environment instead of letting them overlap (#9394)
• fix(credentials): send a conflicting OAuth reconnect back with a reason instead of a 500 (#9391)
• fix(chat): normalize markdown list and heading spacing (#9392)
• fix(isolated-vm): drain worker host deploys without failing in-flight executions (#9389)
• fix(deps): backport zod's leading collection size check so oversized bodies stop validating every element (#9388)
• fix(library): consolidate duplicate sections and FAQs added by repeated refreshes (#9390)
• fix(ci): find the last IndexNow submission through its check run, not the run's job list (#9384)
• fix(deps): update nodemailer to 10.1.7 and undici to 7.30.1 for address-parser and fetch advisories (#9383)

Other Changes

• chore(docker): pin Node and prep images for a read-only root filesystem (#9387)
• docs(library): update best-ai-workflow-builders (#9382)
• docs(library): update ai-agent-examples-by-department-and-industry (#9381)
• docs(library): update agentic-ai-coding-tools-what-they-are-and-how-the-top-options-compare (#9380)
• docs(library): update best-conversational-ai-platforms-crm-helpdesk (#9379)

Contributors

• @waleedlatif1

View changes on GitHub

Features

• feat(library): AI Coding Agents vs Code Completion Tools: Which Should Engineering Teams Choose? (#9375)
• feat(mothership): run scheduled jobs as durable unattended chat turns (#9331)

Improvements

• improvement(docs): build feature-branch docs previews only when docs change (#9370)
• improvement(library): single JSON-LD source, Markdown articles, and IndexNow pings (#9355)
• improvement(library): rank related posts by tag rarity, add backlinks, and widen comparison and pagination links (#9354)
• improvement(auth): encrypt stored integration tokens (#9346)
• improvement(execution): run sandbox workers on a dedicated worker host (#9348)

Bug Fixes

• fix(mothership): hold callerless runs' sim_cli to read API scopes (#9371)
• fix(mcp): answer an unpublishable workflow input on tool updates as validation, and match bare-name alternatives titles (#9367)
• fix(mothership): retry a job dispatch whose usage read failed instead of skipping the run (#9366)
• fix(billing): register the Stripe plugin only when the webhook secret is configured (#9363)
• fix(gmail): clear the history cursor only from a poll that held one (#9364)
• fix(db): count the billing and auth pools in the scaling guide and reset their mock between tests (#9365)
• fix(library): correct vendor facts and remove self-ranking in comparison posts (#9360)
• fix(ci): pick the CodeDeploy action when the app Deploy stage has sidecar actions (#9359)
• fix(execution): close retired remote sandbox workers with a normal closure (#9358)
• fix(egress): retry a gateway CONNECT once on connection failure and log each failed attempt (#9351)
• fix(workflows): accept legacy input field types instead of rejecting them (#9350)
• fix(logs): stop loading execution data and snapshots that reads do not use (#9349)
• fix(db): isolate billing and auth connection pools, release the redeploy-check transaction early, cache integration tool schemas (#9347)
• fix(mothership): close the scheduled-job review gaps (#9345)
• fix(workflows): accept json input type and return 400 for unsupported input types (#9342)
• fix(logs): keep provenance sidecars intact through log PII redaction (#9344)
• fix(gmail): drop an expired history cursor instead of re-saving it every poll (#9343)
• fix(mcp): classify unresolvable and egress-refused server URLs in discovery (#9341)
• fix(models): mark OpenAI gpt-4.1-nano deprecated (#9340)
• fix(workflow-tests): keep the error of a mock that throws (#9333)

Other Changes

• docs(library): update ai-agent-examples-by-department-and-industry (#9374)
• docs(library): update agentic-ai-coding-tools-what-they-are-and-how-the-top-options-compare (#9373)
• docs(library): update best-conversational-ai-platforms-crm-helpdesk (#9372)
• docs(library): update openai-vs-n8n-vs-sim (#9369)
• docs(library): update agentic-ai-coding-tools-what-they-are-and-how-the-top-options-compare (#9368)
• docs(library): update best-conversational-ai-platforms-crm-helpdesk (#9362)
• docs(library): update best-ai-agents-for-lead-enrichment-2026 (#9361)
• docs(library): update langgraph-alternatives (#9357)
• docs(library): update top-ai-assistants-2026 (#9356)
• docs(library): update best-ai-automation-tools-2026 (#9353)
• docs(library): update what-is-an-mcp-server (#9352)
• chore(deps): upgrade Bun to 1.4.4 (#9338)
• docs(library): update ai-agent-examples-by-department-and-industry (#9334)
• docs(library): update how-to-create-an-ai-agent (#9335)
• docs(library): update dify-alternatives (#9336)
• docs(library): update best-ai-agents-support-ticket-triage (#9337)

Contributors

• @waleedlatif1

View changes on GitHub