What's New in Willow
Every capability, improvement and fix shipped to the platform. Governed access for AI agents, updated weekly.
One email per release. Unsubscribe anytime.
API Token Approvals
New API tokens can require admin approval, with dedicated token scopes, a Requests tab, and notifications.
Read the docsAgent Requests with Approval
Users request new background agents, and admins approve or decline them from a dedicated Requests tab.
Read the docsAgent Session Story
A session view that tells the story of each run, with replay, stop, follow-up, rename, archive, and export.
Read the docsGuard Check Flow
Visualizes each guard check: the source, every guard and check that ran, and exactly what failed.
Read the docsLogs Explorer
A new logs explorer with saved views, presets, custom ranges, field-split charts, and a live logs mode.
Read the docsClaude Opus & Sonnet 5.5
Claude Opus 5.5 and Sonnet 5.5 are available for background agents, with Sonnet 5.5 recommended.
Read the docsGit Commit Attribution
Identifies commits written by AI agents, with author, diff stats, and a verified source repository.
Read the docsIsolated External Gateway
Add an isolated external gateway running on a private, internal host to your organization.
Read the docsReadable File Downloads
Files from Slack, Jira, Confluence, and Drive tools come back as readable text instead of base64. Beta.
Read the docsGoogle Tasks Tools
The Google Workspace connector now includes Google Tasks tools for reading and managing task lists.
Read the docsWillow Docs MCP
Search Willow documentation from any AI agent through a new Willow Docs MCP server in the catalog.
Read the docsUniversal MCP Endpoint
One address, mcp.withwillow.ai, routes each user to their org, with a default MCP and sign-up built in.
Read the docsGet Started Onboarding
A guided quick start for new organizations, with an onboarding map and tabs for each capability.
Read the docsPer-MCP Static Headers
Attach fixed HTTP headers to an MCP for background agents, in every auth mode, from the Tools tab.
Read the docsMCP Doctor Tool
A built-in doctor tool in @mcp-s/mcp that diagnoses connectivity, sign-in, and tool list problems.
Read the docsMCP Apps UI Passthrough
The gateway now passes interactive MCP Apps UI from upstream MCP servers through to AI clients.
Read the docsCompany Policy Editor
Publish a company AI policy end users must accept, with Accepted, Outdated, or Pending status per user.
Read the docsCLI Discovery
AI Discovery now finds command-line AI tools, and logs show agent tool inputs and outputs.
Read the docsMicrosoft Teams Agents
Run background agents in Microsoft Teams channels, with end-to-end setup documentation.
Read the docsYour agents are already in the wild.
Give them a Basecamp. Go from AI chaos to AI work, in minutes.
