Glen

Glen is a shared learning layer for AI agents that captures organizational context from agent sessions, Slack, GitHub, tickets, calls, and docs, synthesizing it into a single store. It recalls relevant prior work during a turn and stores new observations in the same round trip, so knowledge gained by one agent is available to the whole organization.

Glen works through a single MCP tool, also accessible via a command line tool, and integrates with clients such as Claude Code, Cursor, and Codex, as well as custom-built agents. It includes a company brain queryable from Slack, any MCP client, or the web app, a PR reviewer and code search that query clones of the agent that made a change, automatic skill distillation and surfacing, and a shared session library with transcripts, live session takeovers, and shared artifacts. Access control runs per observation at recall, with database-level organization isolation and a private mode that keeps sessions out of the store.

Glen is for teams using AI agents across coding and knowledge work. It is available through a design partner program with team onboarding, and the web app, PR reviewer, and Slack bot come with an account.