Changelog

Features

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

Improvements

• improvement(changelog): lead with knowledge base portability

Bug Fixes

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

Other Changes

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

View changes on GitHub

Features

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

Improvements

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

Bug Fixes

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

Other Changes

• docs(library): update ai-agent-examples-by-department-and-industry
• docs(library): update agentic-ai-coding-tools-what-they-are-and-how-the-top-options-compare
• docs(library): update best-conversational-ai-platforms-crm-helpdesk
• docs(library): update openai-vs-n8n-vs-sim
• docs(library): update agentic-ai-coding-tools-what-they-are-and-how-the-top-options-compare
• docs(library): update best-conversational-ai-platforms-crm-helpdesk
• docs(library): update best-ai-agents-for-lead-enrichment-2026
• docs(library): update langgraph-alternatives
• docs(library): update top-ai-assistants-2026
• docs(library): update best-ai-automation-tools-2026
• docs(library): update what-is-an-mcp-server
• chore(deps): upgrade Bun to 1.4.4
• docs(library): update ai-agent-examples-by-department-and-industry
• docs(library): update how-to-create-an-ai-agent
• docs(library): update dify-alternatives
• docs(library): update best-ai-agents-support-ticket-triage

View changes on GitHub

Features

• feat(knowledge): import knowledge base bundles through the API and sim knowledge import
• feat(monitoring): log event-loop lag, stalls, and in-flight route handlers
• feat(ec2): add Amazon EC2 integration
• feat(library): n8n vs Zapier vs Make: Which Automation Platform Is Best in 2026?
• feat(library): LangChain vs LangGraph: What’s the Difference and Which Should You Use?
• feat(library): CrewAI vs LangGraph vs AutoGen: Which Multi-Agent Framework Should You Choose?
• feat(library): OpenClaw vs Claude Cowork: Setup, Pricing, Security, and Scheduling Compared

Improvements

• improvement(insights): shorten the unassigned usage tooltip
• improvement(ci): balance integration jobs and retain documentation build caches
• improvement(executor): size payload compaction bottom-up instead of re-serializing every subtree
• improvement(file-parsers): parse spreadsheets on a worker thread
• improvement(ci): run affected suites and reuse production builds

Bug Fixes

• fix(tests): derive document grant expiry from one timestamp
• fix(ci): isolate desktop shard CLI regressions
• fix(file-parsers): share the event-loop slice across PDF parses and yield inside whole-document passes
• fix(ec2): describe the states each instance state-change tool actually returns
• fix(executor): resume payload compaction on a fresh stack every 256 levels
• fix(telemetry): send browser events to the configured collector and keep logs off the anonymous one
• fix(logs): treat a logs query path as withheld only when its content field was omitted
• fix(logs): mark withheld trace span content so logs query stops reporting it as missing
• fix(logs): yield to the event loop while exporting large trace rows
• fix(parsers): yield the event loop during PDF extraction and CSV row coercion
• fix(logs): cap logs-list pages that carry execution data
• fix(deps): update Next.js to 16.4.1, stop untrusted Office uploads reaching file-type's ASF loop, and patch transitive advisories
• fix(executions): skip archive materialization on status polls that read no output
• fix(logs): skip workflow state on logs query trace reads
• fix(logs): keep trace secret projection from blocking the event loop
• fix(library): correct facts in four new comparison posts

Other Changes

• docs(telemetry): describe the relay's disable gate and the anonymous endpoint's traces-only export
• ci: balance macOS Electron smoke shards by measured duration

View changes on GitHub

Features

• feat(chat): recall sent messages from composer
• feat(library): 6 Best Workato Alternatives for AI Agents in 2026
• feat(library): 6 Best OpenClaw Alternatives for Teams in 2026
• feat(library): 6 Best Lindy Alternatives in 2026 for AI Agents and Automation
• feat(library): 5 Best Claude Cowork Alternatives in 2026 for Shared and Scheduled AI Agents

Improvements

• improvement(emails): refresh the platform update broadcast
• improvement(docs): reduce navigation payload and build work
• improvement(mothership): calmer activity header: wait wording until a call's title streams, and pacing across a run's calls

Bug Fixes

• fix(mothership): hold a tool call's display grouping beside its arguments
• fix(docs): preserve clipboard activation and validate copied content
• fix(chat): validate restored contexts against send schema
• fix(mothership): report a paused Copilot workflow whose cancel lost the race to an unreported terminal run
• fix(mothership): report a Copilot workflow tool whose paused run settles after a resume or cancel
• fix(mothership): let the agent copy files between workspaces without Projects
• fix(mothership): keep display activity out of integration and mcp tool arguments
• fix(deps): pin patched xmldom, axios, follow-redirects, qs, proxy-addr, shell-quote, joi, and grpc-js
• fix(gemini): send the lowest thinking level when thinking is 'none' on Gemini 3

Other Changes

• docs(library): update langgraph-alternatives

View changes on GitHub

Features

• feat(mothership): discover every workspace mcp server without tagging
• feat(mothership): send workspace skill ids and descriptions in the chat inventory
• feat(helpscout): knowledge base connector, auto-registered triggers, 12 more tools, and validation fixes
• feat(helpscout): add Help Scout integration with OAuth and app credentials, selectors, and webhook triggers

Improvements

• improvement(docs): simplify integration page headers
• improvement(docs): refine screenshots, copy, and API interactions
• improvement(insights): clarify unassigned usage and refresh usage charts
• improvement(mothership): rank integration discovery by bm25 and connected services
• improvement(library): merge near-duplicate posts into their stronger pages
• improvement(mothership): show the worker's display copy for a turn error everywhere it is recorded
• improvement(sim-cli): publish each command's path aliases and renamed paths in the command inventory

Bug Fixes

• fix(ci): reset a broken desktop-live dev cache from a passing job
• fix(mothership): bound best-effort mcp discovery and stop retrying a down server every query
• fix(helpscout): require explicit tags to clear, validate custom field entries and positive IDs, and correct trigger and credential copy
• fix(insights): keep paused calendars visible and chart scales finite
• fix(comparisons): Retool self-hosting is on every plan, not Enterprise only
• fix(comparisons): lead Flowise copy with its end of life, qualify Retool self-hosting, and note the Zapier Agents migration
• fix(mothership): keep untagged mcp discovery side-effect free and the catalog resilient
• fix(tools): advertise json and any params without an object type, and name the invalid mcp argument
• fix(mothership): keep integration discovery working when a tagged mcp server fails or lists no tools
• fix(chat): keep previous messages stable when sending

Other Changes

• docs(library): correct vendor facts and Flowise status found in the library audit
• docs(library): rework Zapier alternatives into the lean roundup format
• docs(library): restructure open source AI agents guide around its head term
• docs(library): tighten n8n alternatives and fix its tool count
• docs(library): rework no-code AI agent builders roundup into the lean format
• chore(mothership): sync the Search assistant tool contract from mothership

View changes on GitHub

Features

• feat(mothership): let an agent chat list its owner's other live chats for history recall

Improvements

• improvement(search): return 30 previews by default for a Search MCP query across every source
• improvement(changelog): more relevant Search across sources

Bug Fixes

• fix(tls): distrust the roots behind the compromised eMudhra intermediates
• fix(credentials): declare service-account oauth policy on Snowflake, NetSuite, and Managed Agent tools

Other Changes

• docs(library): update byok-multi-model-ai-agent-builder
• docs(library): update best-ai-agents-for-scheduling-and-calendar-management-in-2026
• docs(library): update can-ai-agents-talk-to-each-other-multi-agent-communication-explained
• docs(library): update apache-2-0-vs-fair-code

View changes on GitHub

Features

• feat(library): Best Retool Alternatives in 2026 for Building AI Agents
• feat(library): Best Dust Alternatives in 2026 for AI Agent Workflows
• feat(library): Best Flowise Alternatives in 2026 for Production AI Agents
• feat(library): Best Langflow Alternatives in 2026 for Production AI Workflows
• feat(library): Best Vellum Alternatives in 2026 for Non-Engineers
• feat(library): Best Pipedream Alternatives in 2026 with Visual Builders

Improvements

• improvement(deps): shared Standard Webhooks verifier, MCP SDK 1.32.1, tiptap 3.31.4 with the prosemirror paste XSS fix
• improvement(comparisons): noindex self-hosted marketing pages, link every comparison, and add Zapier, Retool, Flowise lead answers

Bug Fixes

• fix(comparisons): stop recommending Flowise after its end of life and source its facts from Flowise
• fix(comparisons): Retool lists self-hosting on Enterprise only

Other Changes

• chore(mothership): drop unused worker contract copies
• chore(deps): audit and update dependencies, raise security override floors, collapse duplicate versions

View changes on GitHub

Features

• feat(library): How Do You Connect a Google Drive MCP Server to an AI Agent in Sim?
• feat(library): LLMs vs AI Agents: What Is the Difference?
• feat(greenhouse): connect Harvest v3 with a customer-owned API credential
• feat(zoho-desk): expand help desk operations
• feat(excel): expand workbook operations and align Graph contracts
• feat(greenhouse): add versioned Harvest v3 integration
• feat(cloudflare): expand API coverage and validate operations
• feat(library): 6 Best AWS Bedrock Agents and Vertex AI Agent Builder Alternatives for Easier Deployment
• feat(library): How Do AI Agents Use Tools and APIs?

Improvements

• improvement(async-jobs): warn when Trigger.dev run acceptance is slow
• improvement(db): name chat uploads in one indexed read and index workspace payer lookups
• improvement(landing): search-length titles, model comparison pages, and free model tools
• improvement(mcp): prepare the hosted MCP server for registry and plugin directory listings

Bug Fixes

• fix(execution): precheck the default video target and attribute 4xx resume preprocessing refusals to the run
• fix(cloudflare): require explicit R2 lifecycle replacement rules
• fix(library): point cross-links at the renamed 15 Best Zapier Alternatives guide
• fix(billing): stop free-plan workflow runs at their usage limit mid-run
• fix(greenhouse): align configured agent inputs with API parameters
• fix(execution): classify SDK run aborts and refuse doomed provider and media requests before sending
• fix(api): answer upstream MCP, OAuth, and id-conflict failures with 4xx; keep PII keep-alive past the load balancer
• fix(file): bind copied workspace files to their source by storage key
• fix(landing): correct model version names, fit model page titles, and fix the Zapier post title count
• fix(memory): read memory records and their provenance in one statement
• fix(custom-tools): skip title conflicts when persisting inline tools from a workflow save
• fix(copilot): read sandbox paths in files view and admit missing overwrite targets in multi-file exports
• fix(media): send ElevenLabs TTS text under the field fal requires
• fix(mcp): route deferred transport requests under the client's outbound scope
• fix(account-deletion): list storage keys through the S3 client so Trigger runs can sweep
• fix(streaming): stop writing workflow stream frames after the client cancels
• fix(providers): refuse an empty Responses input before sending instead of retrying a guaranteed 400

Other Changes

• docs(library): update best-conversational-ai-platforms-crm-helpdesk
• docs(library): update n8n-alternatives
• docs(library): update best-ai-agent-platforms-for-enterprise-teams-2026
• docs(library): update best-zapier-alternatives
• docs(library): update agentic-ai-coding-tools-what-they-are-and-how-the-top-options-compare
• docs(library): update ai-agent-examples-by-department-and-industry
• docs(library): update best-conversational-ai-platforms-crm-helpdesk

View changes on GitHub

Features

• feat(library): 6 Best Relevance AI Alternatives for Flexible AI Agents in 2026
• feat(library): How Much Does It Cost to Build and Run an AI Agent?
• feat(library): 6 Best Make.com Alternatives for AI Agents in 2026
• feat(databricks): expand API coverage and preserve paginated results
• feat(oracle-fusion): add core financial business operations
• feat(data-drains): add Microsoft Sentinel ingestion
• feat(project-files): let Chat read and write Project files

Improvements

• improvement(computer-use): roll out computer use per organization
• improvement(landing): serve the homepage as Markdown when clients ask for text/markdown
• improvement(library): refresh license comparison and Zapier alternatives

Bug Fixes

• fix(logs): remove the client log-persistence route that let a run's actor rewrite its log and cost
• fix(resume): require write access to resume a paused run on the legacy route
• fix(auth): end OAuth grants on SCIM address change and add sign out other devices
• fix(copilot): re-derive the turn's secret authority on the in-band tool lane
• fix(executor): read files in Copilot-started runs and replay repeated memory attachments once
• fix(auth): revoke sessions and OAuth grants on password reset
• fix(salesforce): pin instance URL to Salesforce domains and guard update-custom-field egress
• fix(databricks): preserve mutation intent and complete task outputs
• fix(credentials): reject managed OAuth kind mismatches before resolution
• fix(chat): release-audit follow-ups for Project file tabs and sandbox copy parsing
• fix: release-review follow-ups for account removal (workflow-state rejections, stats-row creator, sweep pages)
• fix(chat): treat browser tab metadata as untrusted and bound attachment resolution
• fix(chat): open and close Project file tabs by their owner
• fix(account-deletion): resume the workspace storage sweep from its saved listing position
• fix(search): count two-letter query terms when ranking across sources
• fix(billing): stop account deletion from resetting the free allowance
• fix(account-deletion): durably purge a deleted account's storage without touching live objects
• fix(execution): stop in-flight workflow runs when an account is deleted, removed or banned
• fix(trigger): do not alert on runs whose workspace was deleted mid-run
• fix(execution): end runs quietly when their workspace is deleted mid-run
• fix(trigger): complete async workflow runs whose failure is already recorded instead of failing them
• fix(files): answer 499 instead of 500 when a client disconnects mid-serve

Other Changes

• docs(library): update apache-2-0-vs-fair-code
• docs(library): update best-zapier-alternatives
• chore(ci): remove the flaky-test issue workflow
• revert(search): restore the v0.9.23 Jira text search

View changes on GitHub

Features

• feat(library): How Do You Embed an AI Agent Inside Your SaaS Product?
• feat(library): How Do You Build an AI Agent That Remembers Context Across Multiple Sessions?
• feat(library): 6 Best Microsoft Copilot Studio and Power Automate Alternatives for AI Agents in 2026
• feat(library): 6 Best CrewAI Alternatives in 2026 for Visual and Flexible Agent Building

Improvements

• improvement(search): rank live results across sources by the query terms they hold

Bug Fixes

• fix(search): keep the Jira half-terms summary tier within the all-terms match
• fix(desktop): let computer use type into date and time pickers

View changes on GitHub