Changelog

Features

• feat(demo): auto-confirm demo bookings that came through the verified form and decline the rest

Improvements

• improvement(desktop-browser): let a page finish rendering before observing an action
• improvement(search): shorter Jira any-term budget and a shared batched read budget
• improvement(executor): run the Agent block on the shared model-fallback walk
• improvement(computer-use): allow every app by default, protect sensitive apps, and make per-app approval optional
• improvement(credentials): resolve agent fallback credentials once, Vanta 401 grace, true refusal codes, headless client-credential modes, and registry cleanups

Bug Fixes

• fix(credentials): replay remembered refusals as credential errors, derive the delegable Google services list, and correct a fallback TSDoc
• fix(credentials): offer keyless Google delegation only for services that can be delegated
• fix(aws-marketplace): confirm a container license lapse on a second re-check before switching Enterprise off
• fix(computer-use): refuse Script Editor, Automator and Shortcuts, which run shell commands outside the Terminal control
• fix(search): check the last shortened read window against its budget
• fix(pii): mask texts over spaCy's length limit and drop encoded files from logs
• fix(aws-marketplace): check the container license with CheckoutLicense so AWS's standard buyer policy works
• fix(docker): ship the realtime and migrations images with runtime dependencies only
• fix(desktop): weak-link ScreenCaptureKit so the Computer Use helper loads on macOS 14 and 15
• fix(docs): copy section permalinks and polish outline navigation
• fix(api): match pause points with jsonb_path_exists so a status-filtered run list stays fast

Other Changes

• chore(pii): drop the over-long text example from the mask client's outage comment
• chore(upgrade): drop the stale Cal.com comment on the sales demo link
• chore(analytics): remove Freebuff ad tracking and the Beluga short-link rewrite

View changes on GitHub

Features

• feat(credentials): configurable audience for keyless AWS, Databricks, Snowflake, Microsoft, and Google credentials
• feat(gitlab): add permission CSV exporter
• feat(computer-use): enable computer use by default, governed by a permission-group restriction
• feat(library): How to Build a Shopify AI Agent for Orders and Inventory
• feat(library): AI Agent Platform Vendor Viability: Funding, Adoption, and Staying Power Compared
• feat(aws-marketplace): bill a listing's LLM usage to AWS through a per-period metering tally
• feat(credentials): saved PostgreSQL, MySQL, and SQL Server connections
• feat(credentials): second auth modes for ServiceNow, Snowflake, Infisical, and Google Ads
• feat(knowledge): let the S3 connector sync with a saved AWS credential
• feat(credentials): saved credentials for Amplitude, Kalshi, Splunk, CB Insights, Railway, and Qdrant; fix Kalshi Ed25519 signing
• feat(credentials): Microsoft app credentials for Power BI and Azure Data Explorer, with national clouds
• feat(credentials): keyless Google Workspace domain-wide delegation via IAM signJwt
• feat(credentials): record, show, and explain saved token expiry
• feat(bedrock): run Bedrock models with a saved AWS credential
• feat(credentials): regional and instance hosts for saved LaunchDarkly, LangSmith, New Relic, Square, SendGrid, Hex, and Sentry credentials
• feat(apollo): add optional OAuth connections
• feat(aws-marketplace): sell several listings from one integration through a product catalog

Improvements

• improvement(desktop-browser): click through a nested control that fills the whole target
• improvement(desktop-browser): stable element refs, wait observations, and fewer false refusals
• improvement(aws-marketplace): close billing gaps and simplify the license write paths
• improvement(credentials): keep valid client-credential tokens through refresh outages and pace token retries by failure kind
• improvement(changelog): cover meaningful product updates through v0.9.54

Bug Fixes

• fix(credentials): point the AWS keyless hint at the policy audience, quote the docs client ID, and correct a stale comment
• fix(aws): round-trip typed DynamoDB cursors and allow percentile-only CloudWatch statistics
• fix(desktop-browser): give a wait's closing snapshot a full budget and confirm ref reuse names the same element
• fix(executor): resolve saved model credentials inside each fallback attempt in Router and Evaluator
• fix(integrations): return the Dataverse page size and correct Contacts search outputs
• fix(tests): close three integration-test races that flaked CI
• fix(integrations): final review pass for the Atlassian and AWS sweeps
• fix(integrations): close stale and empty value leaks from the Google and Microsoft sweeps
• fix(desktop-browser): refuse reissued earlier refs and recover from uncommitted ref reuse
• fix(gitlab): require confirmed emails in permission exports
• fix(aws): validate AWS integrations against SDK and API docs
• fix(atlassian): validate Jira, Confluence, JSM, Bitbucket, Trello against vendor docs
• fix(microsoft): validate Microsoft integrations, return dropped fields and expose missing params
• fix(google): close post-merge verification gaps from the Workspace sweep
• fix(google): validate Workspace integrations, return dropped fields and expose missing params
• fix(mothership): back off the transport poll instead of retrying every second
• fix(credentials): Splunk password sign-in, Snowflake key pair in the CLI, Apollo OAuth people search, and one short SailPoint throttle wait
• fix(api): report a finished run's real status once its pauses have settled; show the request id in the CLI

Other Changes

• chore(integrations): tidy two comments left by the Google/Microsoft sweep
• chore(api): describe the settled-pause rule in the run status overlay TSDoc
• revert(library): restore orchestration and governance posts to before #9938 and #9939
• docs(library): remove duplicate questions in orchestration and governance posts
• docs(library): update govern-ai-agents-multiple-teams-enterprise-workspace
• docs(library): update ai-agent-orchestration-frameworks-explained
• docs(library): update enterprise-ai-agent-compliance-deployment-guide

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(aws-marketplace): gate Enterprise on the AWS Marketplace container license and add the EKS install profile
• feat(credentials): audit fixes and non-OAuth auth for CrowdStrike, Datadog, Gong, PlanetScale, SailPoint, ServiceNow, Zendesk, ZoomInfo, Shopify, Atlassian, Linear and Twilio
• feat(api): failing block on polled runs, deployed input schema, paused-runs list, and applied limits
• feat(api): workflow outline, workspace schedules list, clearer v2 errors, and catalog value types
• feat(api): agent identity on v2 meta, scoped CLI pairing keys, and v2 workspace rename
• feat(table): add Columns to Return to the legacy Table block's Query Rows
• feat(library): Best AI Agents for Spreadsheet Analysis in Excel and Google Sheets
• feat(library): Best AI Agents for GitHub Pull Requests, Issue Triage, and Code Review
• feat(library): Best AI Agent Platforms for HR Teams Answering Employee Questions

Improvements

• refactor(credentials): resolve Vanta tokens through the coordinated client-token resolver
• improvement(table): document that a column selection matching no column returns empty row data

Bug Fixes

• fix(credentials): age-gate 403 token eviction and tidy the client-credential paths
• fix(api): attribute a polled run's failure only to the block that caused it
• fix(credentials): match JSON 403s case-insensitively and refuse a malformed ZoomInfo access token
• fix(security): refuse an extra slash after https:// in typed base URLs
• fix(api): release-audit follow-ups for the agent API work
• fix(http): keep unwrapping legacy success envelopes for existing workflows
• fix(selectors): redact email path segments from provider failure logs; correct table timeout guidance
• fix(workflows): refuse operation edits that would overwrite a concurrent change
• fix(chat): bring a restored workflow's chat back online when it is relaunched
• fix(http): stop treating 200 bodies with status and data keys as proxy responses
• fix(tables): show rejected filter errors in the grid and fill imports to the row limit
• fix(errors): actionable table read timeouts, 409 for duplicate custom tools, Instagram revocation, selector failure logs

Other Changes

• chore(deps): tidy comments and override order from the dependency sweep
• docs(api): guide for building and running workflows with the API
• chore(deps): update minor and patch dependencies past the release-age gate
• chore(deps): bump turbo to 2.11.10
• chore(deps): bump AWS SDK to 3.1162.0, OpenTelemetry to 0.223.0/2.12.0, and @daytona/sdk to 0.227.3
• chore(deps): upgrade docx to 9.15.2
• chore(deps): bump mdast-util-to-markdown to 2.2.2
• chore(deps): bump tiptap to 3.32.2 and keep headings that open a list item
• chore(deps): bump knip to 6.44.1
• chore(deps): bump streamdown to 2.7.1
• chore(deps): bump electron-builder to 26.18.0
• chore(deps): upgrade monaco-editor to 0.57.0
• chore(deps): upgrade cronstrue to 3.38.0
• docs(library): update postgres-mcp-server-ai-agents
• docs(library): update best-ai-agents-sales-crm-automation
• chore(api-keys): remove the api-key-read-only flag now that read-only keys are live

View changes on GitHub

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

Features

• feat(credentials): set up keyless credentials from the v2 API, CLI and Sim MCP
• feat(credentials): keyless Snowflake, Microsoft app-only and Google keyless, GitHub and Bitbucket tokens, DocuSign JWT
• feat(mistral): extract PDF images as files and expose table format in Mistral Parser
• feat(api): run a pinned deployment version and expose paused run points in v2
• feat(api-keys): read-only API keys and expiry at creation
• feat(api): workflow trigger health and enablement through the v2 API
• feat(network): honor HTTP_PROXY, HTTPS_PROXY and NO_PROXY for guarded requests, AWS SDK clients and OTLP export
• feat(credentials): saved credentials for 21 token-and-host blocks on one host-aware check factory
• feat(api): manage workspace members through the v2 API
• feat(credentials): saved credentials for 17 more blocks, GitLab tokens, and keyless hardening
• feat(search): manage source permissions through a shared API and CLI
• feat(search): update GitLab permission CSVs through the API
• feat(library): AI Agent Development Company vs Building In House: How to Decide
• feat(library): Best AI Agent Platforms That Support the A2A Protocol in 2026
• feat(library): AI Agent Builders With Built-In Web Search and Research Tools in 2026
• feat(credentials): saved credentials and credential-only block versions, with keyless AWS/Databricks OIDC

Improvements

• improvement(api): release audit follow-ups for v2 member analytics, workflow access middleware and docs
• improvement(search): return a longer live read window when an MCP client asks for more chunks

Bug Fixes

• fix(credentials): correct keyless setup instructions and enforce the saved-credential contract on every credential block
• fix(mistral): say V3 parses an uploaded or referenced PDF, and bound Max Images at 100
• fix(release): host_not_allowed for refused hosts, internal-JWT-only secret release, and final release-audit cleanups
• fix(sim-setup): stop doctor --fix from turning ${VAR} references into literals
• fix(credentials): report a refused host as the user's input, persist normalized account ids, cap Databricks reads
• fix(release): keyless subject uniqueness, NO_PROXY IPv6/CIDR, proxied boot secret, and remaining audit follow-ups
• fix(release): small release-audit cleanups
• fix(release): reuse only full-access delegation keys and treat a blank crawl depth as unset
• fix(security): harden saved-credential, proxy egress, and keyless-mint paths from the release audit
• fix(release): redact deploy keys, check operation-backed tools in the saved-credential contract, and drop SDK copy
• fix(release): run pinned async versions on their own variables, map CLI key-creation errors, show GitLab tokens as the current block
• fix(sim-setup): quote env values containing $ so dotenv, Bun, and Compose don't expand them
• fix(release): cap proxied plain-http responses, count Firecrawl depth from the start URL, and drop stale route cache on upgrade
• fix(firecrawl): complete v2 response and depth compatibility
• fix(cli): bind sim-setup's containers to loopback, fix reset and env quoting, retire the legacy simstudio CLI
• fix(docker): keep ISR and suffix-list caches writable on a read-only root

Other Changes

• test(api-keys): use a neutral database role name in the internal-error fixture
• test(credentials): use a non-token-shaped GitHub fixture so secret scanning stays quiet
• docs(release): correct the Discord allowlist note, Mistral image naming, and the NO_PROXY reference
• docs(skills): name refuseUnsafeTypedUrl in the add-integration token-credential steps
• chore(sdk): use a live release family as the desktop feed's unrelated-release fixture
• chore(sdk): remove the TypeScript and Python SDKs
• chore(deps): take vitest 5.1, shiki 4.6, react-hook-form 7.91, MCP SDK 2.4, and small in-range updates
• chore(deps): upgrade pi to 1.5.0
• ci: bump the actions group across 2 directories with 7 updates

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(docs): polish code rendering and keyboard navigation
• fix(next): align with route-handler framing in 16.4.5 and hash Bun patches into turbo tasks
• fix(trigger): keep each worker's egress gateway address out of the env sync

Other Changes

• chore(deps): move e2b to 2.62.1 and drizzle-kit to 0.31.11
• test(e2e): stop requiring Content-Length on streamed Route Handler downloads
• chore(deps): take the Next.js 16.4.5 security release, patch node-gyp's undici, and refresh in-range patches
• chore(deps): drop the expired Trigger.dev release-age exemption

View changes on GitHub

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

View changes on GitHub