Changelog

Features

• feat(library): Best ChatGPT alternatives for building AI agents and automating workflows
• feat(library): Best AI Workflow Builders for Small Teams in 2026
• feat(forks): opt-in fork sync for new workflows

Improvements

• improvement(search): accept several same-kind native queries per account and cap oversized read limits
• refactor: remove dead code and enforce code quality gates

Bug Fixes

• fix(tables): settle a resume whose pause cannot be saved as failed, and never fail a completed run
• fix(forks): route new workflows through one row builder and harden lineage locking
• fix(logs): keep compacted child span trees shaped as trees
• fix(knowledge): name embedding quota, key, and deadline failures in knowledge search
• fix(knowledge): shrink the member lifecycle budget tests so they stop timing out in CI
• fix(function): report code-placeholder compiler invariants as server failures
• fix(tables): write the cell state when a resumed run throws
• fix(ui): clean up stale workflow state and provider policy handling
• fix(selectors): rebind Copilot principals for nested domain use cases
• fix(tables): match unique JSON values exactly and validate row writes against the live schema

Other Changes

• test(forks): assert the sync-default audit fan-out against real audit_log rows
• docs(library): update best-ai-agents-for-scheduling-and-calendar-management-in-2026
• docs(library): update best-ai-agents-for-executive-assistant-tasks
• docs(library): update apache-2-0-vs-fair-code
• docs(library): update how-to-turn-a-workflow-into-a-reusable-mcp-tool
• docs(library): update sim-vs-dify-open-source-ai-workspace-vs-llm-app-rag-platform
• chore(deps): bump nodemailer to 10.0.10 and patch vulnerable undici copies

View changes on GitHub

Features

• feat(providers): add Kie provider for Claude, GPT, Grok, and Kimi models
• feat(search): add HubSpot CRM member search
• feat(providers): add Claude Sonnet 5.5 and make it the default model
• feat(docs): OAuth scopes section on each integration page
• feat(otter): add Otter.ai integration with triggers and knowledge base connector
• feat(emails): introduce Desktop, CLI, and MCP newsletter
• feat(search): read Drive documents and Gmail conversations
• feat(library): AI Agent Workflow Builders for Multi-Step Tasks: 6-Platform Comparison
• feat(library): How do you build an AI Slackbot without code?

Improvements

• improvement(chat): reuse shared overflow fades on scroll edges
• improvement(tables): bump rows version once at commit and serialize same-value unique writes

Bug Fixes

• fix(chats): align organization shift-click selection with workspaces
• fix(search): preserve Gmail reads when replies disappear
• fix(sim-setup): report Compose db/redis state in status
• fix(desktop): keep built-in browser logins across restarts, faster tab reveal
• fix(providers): retry transient failures on OpenAI Responses and Gemini

Other Changes

• docs(library): update sim-vs-dify-open-source-ai-workspace-vs-llm-app-rag-platform
• docs(library): update best-no-code-ai-agent-builders-2026
• docs(library): update ai-agents-for-marketing-automation
• docs(library): update best-ai-agents-for-slack
• docs(library): update ai-agent-orchestration-frameworks-explained

View changes on GitHub

Features

• feat(chat): add shared find menu
• feat(search): expand live providers and discussion reads
• feat(snowflake): add Cortex Analyst operations to the Snowflake block
• feat(databricks): add Genie agent operations to the Databricks block

Improvements

• refactor(db): require a transaction for advisory lock helpers
• improvement(knowledge): walk connector reconciliation by id so seen stamps stay off the index
• improvement(knowledge): serve date tag filters from an index
• improvement(db): drop unusable date slot and redundant prefix indexes
• improvement(billing): share the enterprise reporting usage read across processes
• improvement(db): tag advisory lock statements by caller
• improvement(knowledge): maintain keyword projections only for search indexes
• improvement(chat): stop reloading the full transcript on every turn and event
• improvement(knowledge): scan processing recovery per connector and prune finished outbox events

Bug Fixes

• fix(knowledge): serve list totals when a request omits the count flag
• fix(search): validate a new provider server before the approval transaction
• fix(knowledge): resume the member resurrection walk from a persisted cursor
• fix(databricks): send workspace tokens only to Databricks workspace hosts
• fix(chat): keep find stable across footnotes and pending chats, and match mention chips next to punctuation
• fix(billing): key the usage email claim on the exact period start and anchor shared sums to their start
• fix(search): let cancelled live reads stop instead of reporting partial coverage
• fix(search): validate Notion search terms before dispatch
• fix(function): omit oversized display code instead of failing the execution
• fix(chat): use hostname brand icons for inline sources
• fix(desktop): wait for native update staging and verify installation
• fix(execution): bound reference and context processing
• fix(api): constrain workflow response headers
• fix(slack-search): stop recovered tool failures from flagging the whole reply
• fix(copilot): bound agent CLI grep matching
• fix(api): prevent caching authenticated workflow variables
• fix(knowledge): count documents through the reader's access and only where counts are shown
• fix(files): stop saving fetched URLs to Files and look names up by the unique index

Other Changes

• test(knowledge): accept either ordered index in the scale window check and bound discussion case text

View changes on GitHub

Bug Fixes

• fix(search): read Confluence pages through the v2 API the Search grant allows
• fix(tabs): preserve readable labels in crowded tab strips
• fix(search): align landing layout and history rows with Home

Other Changes

• docs(library): update best-ai-agent-platforms-for-enterprise-teams-2026
• docs(library): update byok-multi-model-ai-agent-builder
• chore(flags): drop leftover references to the retired async projection flag

View changes on GitHub

Features

• feat(search): refine sources, history and result states
• feat(chat): resolve org-chat mentions in their owner workspace and let chat tag a whole workspace

Improvements

• improvement(search): keep search history on the Search page only
• improvement(search): consolidate knowledge search around live Sim Search and document-decided workspace retrieval
• improvement(chat): nest folders inside their resources in the org resource menu

Bug Fixes

• fix(files): keep the shared markdown parser editor unmounted
• fix(ui): compact crowded tabs and layer native browser overlays
• fix(trigger): exclude integration test files from task discovery
• fix(slack): simplify app creation and setup order

Other Changes

• chore(deps): upgrade Bun to 1.4.2
• chore(tests): consolidate test patterns — shared Vitest base, central @sim/testing mocks, check:test-patterns, lean agent docs
• docs(library): update best-ai-agents-for-executive-assistant-tasks
• docs(library): update best-no-code-ai-agent-builders-2026
• docs(library): update best-ai-agent-platforms-2026
• docs(library): update apache-2-0-vs-fair-code
• docs(library): update best-ai-agents-support-ticket-triage
• chore(tests): remove low-signal tests and consolidate the testing setup

View changes on GitHub

Bug Fixes

• fix(browser): composite background agent tabs, actionable refusals, timed pointer paths
• fix(ci): build the ARM64 app image on a 16 vCPU runner
• fix(ci): build the ARM64 app image on a 16 vCPU runner

View changes on GitHub

Features

• feat(mcp): support OAuth on workflow MCP servers

Bug Fixes

• fix(copilot): clarify knowledge connector authentication

View changes on GitHub

Features

• feat(knowledge): project document ACL and chunk changes asynchronously
• feat(library): Best AI Agent Builders for Human Approval Workflows

Improvements

• improvement(insights): serve org usage from settled segments and redesign the overview

Bug Fixes

• fix(knowledge): address release review follow-ups for ACL pages, projection fill, and sync retries
• fix(insights): keep usage windows on the viewer's local hour
• fix(knowledge): bound every connector-lease ACL transaction to one short page
• fix(jobs): classify transient database failures once for background retries
• fix(knowledge): stop holding the connector table lock across the processing commit
• fix(desktop): set the browser user agent process-wide so Cloudflare Turnstile passes
• fix(db): retry migration lock timeouts within a time budget instead of eight attempts
• fix(file-search): back off indexing retries after database timeouts

Other Changes

• docs(library): update best-ai-agents-sales-crm-automation
• docs(library): update best-ai-agents-for-data-extraction-and-rag-in-2026

View changes on GitHub

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

Features

• feat(jev): add TypeSafe evaluation models to Agent
• feat(models): add GPT-6 Sol and Luna
• feat(providers): add Claude Opus 5.5

Improvements

• improvement(settings): consolidate access requests in settings
• refactor(ui): narrow local style forwarding and preview geometry
• improvement(ui): share expandable home sections
• improvement(ui): share public chat composer actions
• improvement(ui): use chips for table job actions
• refactor(ui): make measurement styles explicit and reuse preview white

Bug Fixes

• fix(knowledge): unschedule a connector whose credential the source rejected and prompt to reconnect
• fix(billing): keep archived selected workspaces in the invitation applied-state check
• fix(oauth): rotate a refresh chain only from the token it started from and size the refresh lease for every provider
• fix(billing): bound every ledger aggregate at the database and size the gate deadline for two
• fix(billing): let a slow ledger read finish instead of blocking execution at the singleflight default
• fix(access): align invitations and request navigation
• fix(organizations): respect inherited access when inviting members
• fix(mcp): list servers without Credential Group access

View changes on GitHub