How is this different from a CLAUDE.md or rules file?

A rules file is something you write and must keep up to date by hand. Glen accumulates codebase knowledge automatically as your agents work, and it is shared org-wide, so every developer's coding agent reads and writes the same growing store.

Do all my developers' coding agents share the same memory?

Yes. Glen memory is org-scoped, so a convention or gotcha one developer's agent records is immediately available to every other developer's coding agent through one tool.

Shared memory for coding agents

A coding agent rediscovers your codebase from scratch every single session. It re-greps for where auth lives, re-learns that you do not use that deprecated helper, re-derives why a module is structured the way it is, and then the session ends and all of it is gone. Tomorrow it asks the same questions. Run two coding agents, or let a teammate's agent touch the same repo, and neither knows what the other already figured out, so they repeat the same exploration and sometimes undo each other's reasoning. Glen, shared memory for AI agents, gives your coding agents one shared memory they read from and write to, so hard-won codebase knowledge stops evaporating between sessions.

Connect Glen as an MCP tool to your coding agents, in Claude Code, Cursor, Cline, Codex, or any MCP client, and each gains a single capability that retrieves what is already known about the codebase and records new facts in one round trip. Before it starts editing, the agent pulls the conventions, the architectural decisions, and the gotchas the team has accumulated, so it works the way your codebase actually works instead of guessing. As it learns something durable, why a workaround exists, which pattern is canonical, what broke last time, it writes that back without you maintaining a docs file by hand.

Because Glen's memory is org-scoped and shared, this is not one agent's private cache. Every developer's coding agent reads and writes the same store, so a convention one agent learned is immediately available to a teammate's agent, and the institutional knowledge of the codebase compounds across the whole team instead of resetting per session and per person. It is framework-agnostic, the agent in your editor and the agent in your terminal share one memory because Glen is a standard MCP server, and the same knowledge is readable from any other MCP client your team uses. Connect once over OAuth or an API key, and your codebase context grows as a byproduct of normal work rather than as documentation nobody keeps current.

FAQ

How is this different from a CLAUDE.md or rules file?: A rules file is something you write and must keep up to date by hand. Glen accumulates codebase knowledge automatically as your agents work, and it is shared org-wide, so every developer's coding agent reads and writes the same growing store.
Do all my developers' coding agents share the same memory?: Yes. Glen memory is org-scoped, so a convention or gotcha one developer's agent records is immediately available to every other developer's coding agent through one tool.