Changelog
Every new feature, improvement, and fix in Sim, the open-source AI workspace, with release notes straight from GitHub.
v0.9.51
Features
- feat(models): add Claude Fable 5.2
- feat(library): Moving AI Agent Prototypes Into Production: A Practical Guide
- feat(library): 10 Self Hosted AI Agent Use Cases for Engineering Teams
- feat(library): AI Agent Platforms for Running Your Own Fine Tuned Models Compared
Improvements
- improvement(search): run Search MCP calls whose arguments have one reading, and log rejected ones
- improvement(search): tell clients a Gmail result shows only its message's opening
- improvement(landing): surface enterprise identity and audit controls where agents read
- improvement(search): refuse a live chunk index before the provider read
Bug Fixes
- fix(landing): render the Enterprise plan note in llms.txt as its own paragraph
Other Changes
- docs(library): update slack-ai-agent-security-permissions-guide
Features
- feat(library): Best AI Agents for Database Change Monitoring and System Event Alerts
Other Changes
- docs(library): update ai-coding-agents-vs-ai-workflow-agents
- docs(library): update ai-personal-assistant-vs-ai-agent-builder
- docs(library): update ai-native-workflow-automation-vs-traditional-automation
- chore(deps): pin @azure/identity 4.14.1 so tedious resolves a patched msal-node
v0.9.49
Improvements
- improvement(search): send repeated account notes and fixed fields once in Search MCP results
Bug Fixes
- fix(search): read Search MCP date filters without an offset as UTC
Features
- feat(video): accept a start frame image for fal.ai video models
- feat(landing): redirect /security and /trust to the Trust Center
- feat(library): How to Standardize MCP Servers Across an Enterprise: Platforms Compared
- feat(library): 8 Best AI Business Process Automation Platforms for Enterprise Operations in 2026
- feat(library): Best AI Agent Platforms for Product Analytics Automation: Mixpanel, Segment, Amplitude, and PostHog
- feat(library): How to Connect an MCP Server to an AI Agent Workflow
Improvements
- improvement(search): shorter live document IDs
- improvement(search): send each Search MCP result fact once
Bug Fixes
- fix(desktop): return account sites from the source-connect e2e server
- fix(sandbox): drop in-flight fetches once the pool gives up or the time budget passes
- fix(oauth): confirm workspace connect links before starting the connection
- fix(otp): make a correct code single-use and never compare a code without expiry
- fix(docs): keep search responsive and contain sidebar scrolling
- fix(desktop): skip the Local Network hint when Sim Cloud is unreachable
- fix(oauth): reject cross-site Instagram workspace launches
- fix(video): read the active start frame input when serializing video settings
- fix(otp): reserve a verification attempt atomically before comparing the code
- fix(desktop): explain macOS Local Network blocks for private self-hosted servers
- fix(file-parsers): reject ZIP archives JSZip reads from outside the guarded region
- fix(sandbox): cap task timer bridges and dispose isolates on host timeout
- fix(oauth): reject cross-site workspace OAuth launches
- fix(execute): derive credential actor from auth type, not body isClientSession
Other Changes
- ci(npm): publish every npm package through trusted publishing
- docs(library): update best-ai-agent-platforms-for-enterprise-teams-2026
- docs(library): update best-hubspot-alternatives-ai-marketing-automation-2026
- docs(library): update how-to-build-ai-slackbot-without-code
- docs(zoom): clarify account selection and separate search connections
- chore(execute): tidy credential-actor tests and comment
- docs(library): update best-ai-agents-for-executive-assistant-tasks
Features
- feat(media): add Higgsfield to video and image generators, allow typed models
- feat(higgsfield): add Higgsfield integration
- feat(search): add Hex project search and notebook reads
Improvements
- improvement(integrations): show the sites an Atlassian account reaches
Bug Fixes
- fix(oauth): add Atlassian sites and Notion workspaces on the same login
- fix(search): recheck Hex workspace after notebook reads
- fix(integrations): use Flow and Jama brand icons
- fix(tables): keep leading-zero codes as text when inferring CSV column types
v0.9.46
Bug Fixes
- fix(billing): settle hosted fal generations to their exact cost
- fix(mothership): answer a Chat workflow tool paused for a person with its resume link
- fix(billing): refuse Stripe checkout for AWS Marketplace-billed customers
Features
- feat(providers): add Thinking Machines (Inkling) provider
- feat(library): Best AI Agents for Microsoft 365: Outlook, Teams, SharePoint, and Excel
- feat(library): Best AI Agents for Content Operations: Repurposing, Fact Checking, and News Monitoring
- feat(library): AI Agent Builders for Claude, Gemini, GPT, and Open Models Compared
- feat(library): Best AI Agents for Sales Call Summaries and CRM Updates in 2026
- feat(library): 6 Best AI Workflow Tools for Google Drive Document Automation in 2026
- feat(billing): AWS Marketplace SaaS contract registration and entitlement verification
- feat(gauge): add Gauge Chat integration
- feat(integrations): add Jama and Flow workflows and live Search
- feat(search): connect multiple accounts per integration
- feat(library): The 7 Best Free AI Agent Builders in 2026
- feat(library): Best LLM Orchestration Tools for Multi-Provider Call Chaining in 2026
- feat(library): 6 Best Tines Alternatives in 2026 for Teams Beyond SecOps
- feat(library): How Do You Connect Salesforce’s Hosted MCP Server to an AI Agent in Sim?
Improvements
- improvement(search): steer Confluence native queries to siteSearch for topic lookups
- improvement(logs): move correlation ids from message tags into metadata and enforce it
- improvement(logs): log each admission refusal once per window and demote per-request success chatter to debug
- improvement(org): move the People page onto the new design system
- improvement(geo): clarify self-hosting and compliance for answer engines and agents
- improvement(org): move the organization view onto the new design system
- improvement(enrichments): consolidate lookups through Firecrawl
Bug Fixes
- fix(release): include the AWS Marketplace checkout guard
- fix(search): preserve provider guidance for personal connections
- fix(billing): keep the Marketplace lookup retry stable across renders
- fix(billing): refuse Stripe checkout for AWS Marketplace-billed customers
- fix(logs): log the webhook poll id and emitted event id under their own names
- fix(org): drop leftovers of the reverted search history source removal
- fix(billing): withhold hosted Fal.ai video generation from free-plan payers
- fix(docs): align navigation and collapse integration fields
- fix(search): preserve query attribution on fused results
- fix(library): describe self-hosted Enterprise features accurately in comparison posts
- fix(library): use the renamed self-hosted pricing row in comparison posts
- fix(assistant-actions): read a run that closes mid-review as nothing to review
- fix(desktop): route organization members to their organization and settings
- fix(org): drop the removed shortcuts slot from the landing test
- fix(falai): bound pricing lookups and preserve result receipts
Other Changes
- chore(release): include personal Search connection guidance
- chore(release): merge staging into the v0.9.45 release
- chore(release): merge staging into the v0.9.45 release
- chore(agents): trim root CLAUDE.md by ablation; AGENTS.md links to it
- chore(contracts): sync the worker's steer disposition read contract
- revert(org): hold the org view and People redesigns for #9687
- chore(contracts): sync the worker's steering v2 contract
- docs(library): update best-ai-agents-for-data-extraction-and-rag-in-2026
- docs(library): update ai-agents-in-procurement
- docs(library): update what-is-human-in-the-loop-in-ai-agents
v0.9.44
Features
- feat(library): How Do You Build a QuickBooks AI Agent for Month End Close?
- feat(library): How Do You Build a Notion AI Agent for Pages and Databases?
- feat(library): How Do You Build an Airtable AI Agent for Data Entry and Reporting?
Improvements
- improvement(links): make go/ and sim/ work with no setup
- improvement(comparisons): render every comparison question section through one component
- improvement(extension): allow more pages within a shared content budget
- improvement(comparisons): answer Copilot Studio buyer questions in place with an at-a-glance table
- improvement(extension): simplify page access and enable voice input
- improvement(links): open go/ links alongside sim/ and call the page Links
- improvement(ci): parallelize desktop release verification and packaging
- improvement(ci): avoid redundant browser dependency updates
Bug Fixes
- fix(release): include extension sign-in recovery
- fix(extension): keep sign-in recovery available
- fix(desktop): download Bun fresh in the signed release build
- fix(links): keep long searches within the list limit
- fix(desktop): keep signed release builds off shared caches
- fix(search): describe connection availability accurately
- fix(logs): keep checked trace content at every depth and label live projection warnings
- fix(search): keep Confluence searchable for colon queries and say when nothing was searched
- fix(voice): stop microphone capture during startup
- fix(ci): pause search clock without wall-time races
- fix(logs): keep retryable resume refusals internal and log deploy-path config rejections at warn
- fix(logs): name display fields and real span counts in trace projection warnings
- fix(files): let Mothership tool runs write workspace files
- fix(logs): keep checked trace content when the display projection budget runs out
- fix(ci): distinguish handled search redirects from runtime errors
Other Changes
- chore(release): merge staging into the v0.9.44 release
- docs(library): update best-ai-agent-builders-with-mcp-support
- docs(library): update best-marketing-automation-platforms-ai-workflows-2026
- docs(library): update best-gumloop-alternatives-in-2026
Features
- feat(video-generator): refresh native provider catalogs and request protocols
- feat(image-generator): refresh provider model catalogs and validate requests
- feat(extension): improve page capture, attachments, and browser shortcuts
- feat(sim-links): create or copy a Sim link from search results and file shares
- feat(people): add a synced organization chart
- feat(video): expand the fal text-to-video model catalog
- feat(library): 5 Best UiPath Alternatives for AI Agents in 2026
- feat(library): Fastest AI Agent Platforms for Time to Value and Company Rollout in 2026
- feat(library): How Do You Build a Snowflake or BigQuery AI Agent That Reports to Slack?
Improvements
- improvement(sim-links): drop the leftover fragment around the source actions menu
- improvement(ci): run WebKit acceptance in pinned browser containers
- improvement(ci): reuse PR checkout mirrors for integration jobs
- improvement(cleanup): tighten the agent-harness changes from the latest audit
- improvement(search): cite public web results like connected sources in Search Assistant
- improvement(cli): recovery guidance on every parse error and body-shape refusal
- improvement(ci): partition extension acceptance by fixture lifecycle
- improvement(ci): validate pushes independently of deployments
- improvement(ci): avoid unrelated asset checks and browser downloads
- improvement(workflows): lint the manual-run entry check and surface lint findings in apply warnings
- improvement(executor): make run-from-block refusals actionable
- improvement(agent-cli): skip non-text files in grep instead of marking the search incomplete
- improvement(landing): re-encode embedded brand logos to cut comparison page weight
Bug Fixes
- fix(billing): bound the enterprise preflight's workflow run count to the preview budget
- fix(trigger): keep secret text out of env sync errors and refuse an unknown sync target
- fix(file-search): fill dispatch capacity in selection order and keep reaped claims ahead of the rotation
- fix(extension): preserve capture focus and Sim Link routes
- fix(media): preserve video settings and Google URI delivery
- fix(people): harden paused SCIM sync and directory inputs
- fix(file-search): claim a file write that lands on an owner the backfill already queued
- fix(cli): match a dash-led id in linear time
- fix(workflows): charge an operation whose new edge the final scope check drops
- fix(resume): carry the admitted resume run's ID on outer resume failure logs
- fix(webhooks): answer missing trigger configuration with 400 and stop logging refused resumes as errors
- fix(extension): preserve composer focus and pending page reads
- fix(sim-links): open the new-link form above the sources popover and suggest hyphenated names
- fix(file-search): claim owners woken by a file write ahead of backfill reconciliation
- fix(release): keep Entra sync running through mid-read drift and tighten release-audit findings
- fix(workflows): keep the Run-button lint finding advisory and point at lint from apply warnings
- fix(ci): attribute pull requests against the tested target parent
- fix(api): classify transient database reads and client-caused failures instead of answering 500
- fix(mothership): surface the real run error for browser-settled workflow tools
- fix(files): accept files/ VFS paths in v2 file download and sim files get
- fix(admin): stop counting workflow runs on dashboard list pages
- fix(search): remember the last selected view
Other Changes
- chore(release): merge staging into the v0.9.43 release
- chore(release): include Chat extension review fixes
- chore(release): merge staging into the v0.9.43 release
- chore(release): include People sync review fixes
- chore(release): merge staging into the v0.9.43 release
- chore(release): merge staging into the v0.9.43 release
- chore(release): merge staging into the v0.9.43 release
- chore(release): merge staging into the v0.9.43 release
- chore(release): merge staging into the v0.9.43 release
- test(files): prove files/ VFS path downloads over real HTTP instead of a mocked route call
- chore(deps): clear reachable advisories with in-range security releases
- chore(trigger): sync Trigger.dev env vars from the app's Secrets Manager secret, gated per environment
- docs(library): update best-ai-agent-platforms-for-connecting-your-existing-tools
- docs(library): update best-ai-agent-builders-for-human-approval-workflows
- docs(library): update sim-vs-dify-open-source-ai-workspace-vs-llm-app-rag-platform
- ci: bump the actions group across 1 directory with 7 updates
helm-chart-1.11.12
A Helm chart for Sim - the open-source AI workspace where teams build, deploy, and manage AI agents

