Changelog

Features

• feat(memory): authorize scoped note access and durable cleanup
• feat(library): How Do You Build a WhatsApp AI Agent With Sim?
• feat(library): 5 Best Discord AI Bot Builders for Community Automation in 2026
• feat(library): 6 Best AI Agent Options for Ecommerce Operations in 2026
• feat(library): Best AI Agents for Querying SQL Databases in Plain English

Improvements

• improvement(search): follow Jira summaries holding every term with those holding half of them
• improvement(search): fall back to Jira summaries holding half the terms of a longer query
• improvement(search): lead an overflowing Jira all-terms match with issues whose summary holds every term
• improvement(seo): lead comparison pages with a direct answer, re-verify competitor pricing, and rebuild the GitHub, Shopify, and Snowflake integration pages
• improvement(seo): list the v2 API as a catalog item and answer markdown 404s
• improvement(changelog): expand release coverage and refine content styling
• improvement(browser): accept observe on every page action

Bug Fixes

• fix(desktop): tighten computer use input guards for text, tabs, menus and AX failures
• fix(copilot): load workflow chat history through the windowed chat routes
• fix(mothership): keep a started call's result after Stop, and let a blocked owner stop their run
• fix(file-search): stop requeueing skipped artifacts, and fix public share, live-doc persist and cleanup edge cases
• fix(slack): refuse binding a Search bot to a different app or workspace than members
• fix(billing): release the reporting prefix lease once its prefix is stored
• fix(scim): push Stop for runs a directory sync ends, after its transaction commits
• fix(media): bill a completed fal generation even when a Stop ends its download
• fix(mothership): hold desktop leases, the resume wait and first claims to the call deadline
• fix(mothership): serialize headless run admission with chat soft delete
• fix(memory): read note access from an unlocked snapshot and defer unavailable cleanup
• fix(seo): answer Markdown OPTIONS on unknown paths and list every accepted knowledge-base file type
• fix(desktop): count only UI-opening AX actions as delivered on timeout, drop unused contract target
• fix(content): keep stored provenance consistent
• fix(billing): answer an in-flight retry at once, and tell Chat what provenance withheld
• fix(desktop): accept common computer calls and let input reach attached sheets
• fix(mothership): start no tool call with less than a minute of its deadline left
• fix(billing): re-read the usage gate often once a payer nears its limit
• fix(mothership): bound every handed tool call by the deadline its frame names
• fix(mothership): stop live runs when their chat, owner or access is removed
• fix(desktop): make computer use act on the app the user sees
• fix(mothership): parse each checkpoint control request on its own and declare accepted kinds
• fix(db): throttle autovacuum on the largest tables so vacuum WAL bursts stop stalling commits

Other Changes

• docs(library): update best-ai-agent-platforms-for-enterprise-teams-2026
• docs(library): update best-ai-agents-for-regulated-industry-workflows-healthcare-legal-procurement
• docs(library): update langgraph-alternatives
• docs(library): update n8n-alternatives
• docs(library): update byok-multi-model-ai-agent-builder
• chore(mothership): sync the computer-use contract copies

View changes on GitHub

Features

• feat(project-files): add public sharing and protected delivery
• feat(project-files): add cross-owner copy and bulk transfers
• feat(project-files): add rendered delivery and indexed search

Improvements

• improvement(api): publish an RFC 9727 API catalog and stop padding tables in docs markdown

Bug Fixes

• fix(chat): preserve panel resize gestures
• fix(mothership): take over runs whose controller died and re-run their interrupted tool
• fix(files): restore workspace read-path behavior after rendered delivery
• fix(pii): load models only in workers and recycle them before memory runs out
• fix(project-files): requeue rendered-artifact search after commit and trim Project read transactions
• fix(contracts): drop unused project file search query exports

Other Changes

• chore(ci): tag production promotions with release- for rollback retention

View changes on GitHub

Features

• feat(project-files): add core storage and realtime operations
• feat(library): 7 Best AI Agents for Recruiting Teams in 2026: Screening, Scheduling, ATS, and Outreach
• feat(library): 7 Best Voice AI Agent Builders for Phone Support in 2026
• feat(library): Best AI Agents for IT Helpdesk Teams: 6 Platforms Compared
• feat(library): Best AI Agent Platforms for Agencies and Consultants Managing Multiple Clients
• feat: establish shared file ownership foundation
• feat(library): How to Give AI Agents Live Web Search with the You.com API in Sim
• feat(changelog): cover October 13 product releases
• feat(mothership): add super-user Plan and knowledge graph settings with benchmark tooling

Improvements

• improvement(docs): standardize integration copy and generated references
• improvement(naming): rename sandbox-tasks to sandboxes and standardize file names
• improvement(ci): track flaky checks daily and fix the live flakes
• improvement(mcp): advertise server metadata, output schemas, and clearer tool docs
• improvement(docs): standardize copy and responsive navigation
• refactor(files): unify mutation accounting and retention
• improvement(api): declare OAuth scopes in OpenAPI, JSON 404s for unknown API paths, and root MCP resource metadata
• improvement(audits): ratchet relative imports, literal text sizes, static inline styles, db access in API routes, and bare route handlers
• improvement(seo): fix soft 404s, per-integration freshness, and docs markdown for AI crawlers
• improvement(mothership): load chat transcripts as windows from a stored UI projection
• improvement(search): rank an overflowing Confluence all-terms match with relevance search
• improvement(workflow): scope the editor by props instead of the URL

Bug Fixes

• fix(project-files): hold the Project and permission-group locks shared for non-charging file operations
• fix(mothership): drop queued tab adds the server already removed
• fix(file-doc): stop the permission resend loop and restore the 10 MB proxy body limit
• fix(projects): gate the Project list on the projects rollout flag
• fix(mothership): surface actionable workbench file read failures
• fix(logging): log expected outcomes below ERROR and acknowledge a lost workflow-result race
• fix(byok): preserve credential ownership on lookup failures
• fix(appconfig): share the polled snapshot with Trigger.dev runs
• fix(api): map MCP reauth redirect, deleted-account visit, and task service outage to non-500 statuses
• fix(folders): run reorder lock checks on the reorder transaction
• fix(resume): retry automatic resumes after read deadlines and an unreadable usage ledger
• fix(uploads): sign transfers for in-process v2 callers that carry no network origin
• fix(knowledge): serialize chunk date tags on the internal chunk read
• fix(resume): answer a sync v2 resume whose run fails with the failed run, not a 500
• fix(docs): preserve negotiated response cache headers
• fix(docs): correct delivery headers and preview details
• fix(file-search): send legacy workspace index payloads, isolate realtime test DB, best-effort post-upload Project check
• fix(files): cancel resumed runs across replicas, unify anchored file edits, and harden workspace, integration, and presence paths
• fix(content): preserve canonical provenance across reads
• fix(realtime): share file editor presence across replicas
• fix(uploads): drop the stale reconciliation-retry claim from the upload-parts cleanup log
• fix(files): keep file search and compiled-doc reads off write locks
• fix(projects): fence file writers with a shared Project lock
• fix(file-search): find an owner's stale dependencies from its dependency rows
• fix(projects): sweep only retired Project prefixes
• fix(mothership): admit stored content to Chat per item through one gate
• fix(docs): return 404 for unknown docs paths and derive integration counts from the catalog
• fix(mothership): grant gated entitlements to organization Build chats
• fix(settings): add platform admin tabs to organization view
• fix(empty-states): keep const assertions on changelog release constants

Other Changes

• ci: bump the actions group across 2 directories with 5 updates
• docs(library): update best-ai-agents-invoice-processing-finance-reconciliation-2026
• chore(files): tighten #8609 follow-ups
• docs(library): update what-is-an-ai-agent-definition-how-it-works-and-examples
• docs(library): update can-ai-agents-talk-to-each-other-multi-agent-communication-explained
• docs(library): update best-ai-agents-for-customer-support-automation
• docs(library): update best-ai-agents-for-data-extraction-and-rag-in-2026
• docs(library): update sim-vs-dedicated-chatbot-builders
• chore(legal): update privacy and cookie policies
• chore(docs): remove the academy

View changes on GitHub

Features

• feat(fivetran): add connector setup, schema config, group, and destination operations
• feat(fivetran): add Fivetran integration with saved credential, selectors, and webhook triggers
• feat(empty-states): add tests and changelog empty-state graphics
• feat(models): add Gemini 4 Argon

Improvements

• improvement(fivetran): use shared filterUndefined and readonly constant lists
• improvement(search): require half the query's terms in Jira partial matches
• improvement(search): bounded Jira fallback, organization search through integrations page connections
• improvement(search): read up to 10 documents per MCP read_document call
• perf(tables): keep the batch unique check on per-value GIN probes
• improvement(chat): coalesce replayed transcript reloads and narrow text-only message reads
• perf(billing): sum reporting-window usage from a shared prefix plus a short tail
• improvement(knowledge): finish Confluence passes past unreadable attachments, rank unauthenticated keyword search in one statement
• improvement(logs): group dashboard stats by workflow id and join names once per group
• improvement(db): disable fastupdate on the table-rows GIN and cover level in the logs activity index

Bug Fixes

• fix(slack): serialize Search bot removal with member setup, refuse a custom row on the shared app ID
• fix(chat): replace a second stuck transcript reload, and close the cleanup pool in its test
• fix(search): fold -ie plurals in the Jira partial-match term key
• fix(slack): serialize member Slack setup with bot installs and explain blocked setup
• fix(db): wave 1 performance follow-ups from re-audit
• fix(knowledge): bill Gemini spend for image reads that fail the pass
• fix(slack): connect Slack member accounts without the Sim Search bot
• fix(files): lock the workspace row before inserting a workspace file

Other Changes

• ci(e2e): skip the duplicate type pass and source maps in the mobile build
• ci(desktop): run the live suite's lease-lapse test on its own runner

View changes on GitHub

Features

• feat(knowledge): index Confluence image attachments
• feat(projects): enforce Project membership and retire the connector
• feat(library): 6 Best AI Agents for Invoice Processing and Finance Reconciliation in 2026
• feat(library): Best AI Agents for Web Scraping and Market Research in 2026

Improvements

• improvement(search): Confluence and Jira partial-match recall, passage previews, batched MCP reads
• improvement(db): cut hot-path connector count and rate-limit lock costs
• improvement(perf): bound trash, file-path and attachment reads
• improvement(db): disable embedding FTS fastupdate and index delete-action foreign keys
• improvement(changelog): add meaningful release coverage

Bug Fixes

• fix(knowledge): key vision checkpoints by title and keep excluded images off while disabled
• fix(lifecycle): retain shared resources when creators leave
• fix(recently-deleted): dim the list and disable Restore while a new view loads
• fix(db): optimize Project backfill and repair archived workflows
• fix(recently-deleted): keep previous-criteria rows inert while the new list loads
• fix(ui): simplify responsive layouts and compact workflow controls
• fix(auth): rate-limit socket tokens per user, with an IP backstop
• fix(ci): remove Project image digest migration gate
• fix(emcn): keep ongoing status toasts behind actionable toasts
• fix(realtime): end a failed socket-token attempt promptly and honor Retry-After
• fix(e2e): read the visible workspace shell while a streamed page arrives
• fix(desktop-e2e): remove root causes of four retry-masked desktop flakes
• fix(e2e): connect the mobile suite to realtime and give dev-server round trips the server budget
• fix(tests): cross the large-array threshold with 64 wide rows instead of 120k
• fix(enrichments): reduce cost, increase accuracy
• fix(workflows): hold the v2 sync run result until its execution log is finalized

Other Changes

• ci: ratchet type errors in apps/sim test files
• chore(tests): ratchet real wall-clock waits in unit tests
• test(knowledge): cover member access withdrawal on a lost provider scope
• chore(storage): drop resetStorageMethod, whose last caller was a removed quarantined suite
• chore(tests): remove quarantined integration suites
• ci(e2e): temporarily skip tablet workflow-controls mobile check
• ci: run the mobile e2e suite on a production build, one job per browser
• ci: scope pull request checks to what the diff can affect
• chore(changelog): document twice-weekly editorial reviews
• chore(ci): fail closed on missing test Redis in CI and skip unused pg16 service
• chore(db): drop retired storage whose readers have shipped

View changes on GitHub

Features

• feat(changelog): publish the Nebius model update
• feat(library): add Stripe billing and failed-payment agent guide
• feat(library): add Stripe payment automation and alert guide
• feat(library): How Do You Build a HubSpot AI Agent for Lead Enrichment and Follow-Up?
• feat(library): How Do You Build an AI Agent That Answers Questions From Confluence?
• feat(library): AI Agent Version Control and Rollback: Platforms, Process, and Enterprise Checklist
• feat(library): How Do You Control AI Agent Spend Across Teams?

Improvements

• perf(tables): read a sorted view's order once per drain and project selected columns in SQL
• perf(mothership): stop re-reading and over-serving chat transcripts
• perf(logs): order public log pages on an indexed column and retire a superseded document index

Bug Fixes

• fix(chat): name stackless error subclasses and compact stored unfinished tool params
• fix(google-chat): avoid unchanged transcript reindexing
• fix(files): refuse an unknown file per read instead of blanking the Chat tool result
• fix(sandbox): name a failed Function's error from its headline, not an inspected property
• fix(google-chat): preserve complete bounded syncs

Other Changes

• docs(changelog): record completed editorial and media review
• docs(library): update what-is-an-agentic-workflow

View changes on GitHub

Features

• feat(library): How Do You Build a HubSpot AI Agent for Lead Enrichment and Follow-Up?
• feat(library): How to Build a Pipedrive CRM AI Agent for Deal Follow-Ups
• feat(projects): move Project membership to the workspace column

Improvements

• perf(tables): seek keyset row pages instead of scanning from the first key
• perf(auth): stop blocking requests on best-effort last-used timestamp writes
• perf(knowledge): lock the knowledge base row FOR NO KEY UPDATE on document writes
• perf(files): resolve name references through an index instead of listing the workspace
• perf(search): fetch Lucid page regions concurrently on a live read
• perf(search): search up to eight live accounts at once

Bug Fixes

• fix(search): return actionable live read errors and drop unreadable Confluence matches
• fix(file-parsers): cap untrusted number formats and make time-format check linear
• fix(realtime): scope batch-add upserts to the caller's workflow
• fix(inbox): scope reply threading to the receiving workspace
• fix(app): keep square logos out of loading screens

Other Changes

• docs(library): update ai-agent-observability
• docs(library): update what-is-an-mcp-server
• docs(library): update best-ai-agents-sales-crm-automation
• docs(library): update best-ai-agents-for-customer-support-automation
• docs(library): update best-ai-agents-for-regulated-industry-workflows-healthcare-legal-procurement

View changes on GitHub

Features

• feat(changelog): workspace changelog of releases with markdown bodies and the Sim changelog tool
• feat(library): AI Agents for Asana and monday.com Task Automation: Platform and Workflow Guide
• feat(sim-cli): publish which CLI commands Mothership is refused, from the route admission rule
• feat(shopify): expand operations and refresh expiring credentials
• feat(checkr): add Checkr integration with tools, block, and webhook triggers
• feat(tests): workflow tests as a workspace resource
• feat(library): Enterprise AI Agent Compliance and Deployment Guide: HIPAA, GDPR, Data Residency, On-Prem, and VPC
• feat(changelog): publish curated product updates
• feat(analytics): attribute sign-ups and demo requests to their acquisition source
• feat(dashboards): add empty-state graphic for the dashboard page
• feat(oracledb): add Oracle Database integration
• feat(library): 6 Best AI Agent Marketplaces and Template Libraries in 2026
• feat(library): How Do You Build a Salesforce AI Lead-Scoring Agent With Sim?
• feat(oracle-epm): add validated integration foundation
• feat(oracle-fusion): add validated integration foundation
• feat(oci): add validated native integration foundation
• feat(integrations): add saved Ramp and Vanta credentials

Improvements

• improvement(ci): fix the stripe-sync flake, stop helm cancelling a publish, attest via actions/attest
• improvement(app): show branded wordmark while loading
• improvement(logs): log each execution failure once at its owning boundary
• improvement(sim-cli): commands reach the API only through the operations they declare
• improvement(changelog): refresh media and released product updates
• improvement(logs): cut recurring production log noise and fix the bugs behind it
• improvement(diff): refine embedded diff presentation
• improvement(editor): refine previews and compact formatting controls
• improvement(mobile): refine responsive chat and workspace layouts
• improvement(desktop): make file consent clearer and easier to control
• improvement(emcn): refine shared menu and picker styling

Bug Fixes

• fix(chat): keep composer drafts across hydration
• fix(chat): contain mobile composer scrolling
• fix(tools): stop provider timeout params from becoming the request deadline
• fix(docs): render the FAQ JSON-LD as a native script tag, and forbid data-typed next/script
• fix(webhooks): stop admission-refusal retry loops and per-retry log writes
• fix(oauth): record revoked refresh tokens durably and surface them as reconnect
• fix(billing): let operators resolve billing dead letters and stop re-paging old ones
• fix(pii): stop one half-emoji from failing log redaction
• fix(sim-cli): the command inventory describes the operation each command runs
• fix(files): refuse Chat current-file reads no provenance observer records
• fix(deploy): return deployed-version lint as its own field, opt-in and time-bounded
• fix(sim-cli): an unknown option on a command with no options says so
• fix(mothership): show the copy control for API keys created in organization chat
• fix(tests): purge deleted tests, bill their storage, and record versions as loaded
• fix(credentials): allow encoded Oracle connection payloads
• fix(desktop): open Folder Access from a focused utility window
• fix(lint): describe unquoted JSON references as failing only when the text is not itself JSON
• fix(sim-cli): an unknown option lists the command's own flags
• fix(files): refuse delegated version reads no provenance observer records
• fix(mothership): let Chat share credentials and request access as the delegating member
• fix(mothership): let Chat read its own permission config and file version history
• fix(workflows): stop flagging whole-field references in JSON fields and correct the quoting advice
• fix(settings): decouple access settings from usage analytics
• fix(invitations): let Chat invite teammates through the workspace invitations command
• fix(desktop): require folder permission for native file tools
• fix(providers): remove default agent tool-call iteration cap
• fix(mcp): add directory verification compatibility
• fix(mcp): declare request bodies and annotation titles
• fix(desktop-browser): ask the user before a page's alert, confirm, or leave-site prompt is answered
• fix(tables): drop the per-table row-order lock from inserts
• fix(tables): log row writes to the change log instead of locking the table definition row
• fix(google-ads): align reporting with current API access
• fix(db): retire unused keyword search projections
• fix(billing): acknowledge Stripe webhooks whose subscription references a missing organization
• fix(files): support arbitrary uploads and in-document links
• fix(workflows): lint unquoted string references in JSON fields and surface lint on deploy

Other Changes

• chore(audits): check raw-route parseRequest contracts and enforce surface-neutral application imports
• chore(tests): tighten check:test-patterns and shrink its baseline
• docs(agents): align agent guidance with code and enforced checks
• chore(naming): rename 41 baselined files to the file-name convention
• chore(audits): name generated-artifact checks check: so run-audits discovers them
• chore(api): give colliding error helpers domain names, drop VFS pass-through wrappers
• chore(audits): drop dead ESLint directives and redundant check:dead-code; fix the PPTX blend-group call they hid
• chore(google-ads): remove synthetic E2E suite
• docs(library): update best-ai-agent-marketplaces-template-libraries
• docs(library): update best-ai-automation-tools-2026
• docs(library): update best-ai-agents-for-scheduling-and-calendar-management-in-2026
• docs(library): update ai-agents-vs-rpa
• docs(library): update best-ai-agent-platforms-2026
• docs(library): update how-to-create-an-ai-agent
• test(credentials): cover Chat's OAuth-only, admin-demotion and integrations guards on credential sharing
• docs(library): update best-zapier-alternatives
• docs(library): update open-source-ai-agent-platforms
• docs(library): update best-ai-agents-sales-crm-automation
• chore(integrations): refresh catalog and Microsoft branding
• chore(config): retire unused env vars and fully-rolled-out feature flags
• docs(library): update best-ai-agents-support-ticket-triage

View changes on GitHub

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

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