claude-code-agent-starter / agent-os
CLAUDE.md1 starsChanged 4 months agoMITView on GitHub
In this repositoryOne other file it gives its agents.
- templatesSkill
The file
We split this CLAUDE.md into the rules it gives an agent
1 rule. About 0 are common, 0 are distinctive, and 1 is very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈0
Distinctive · estimate
This author's own choices. Worth a look.
1
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Very specific: line 10
Skills: when I describe the same process twice, propose saving it as a skill in agent-os/methods/ (when to use, steps, rules, what done …
≈ “Common” is an estimate: at least 1 in 5 of the 26,980 CLAUDE.md files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
Agent operating system
<!-- Copy this block into your CLAUDE.md (project level, or ~/.claude/CLAUDE.md for everywhere).
Adjust the paths if you placed agent-os/ somewhere else. -->How to work in this system
- Start of work: load
agent-os/TOOLKIT.md. It is the map of every method, tool, and rule. Then loadagent-os/memory/MEMORY.md(the memory index) and open only the memories relevant to the task. - Memories: when you learn something durable (a preference of mine, a decision and its reason, the state of a project), write it to
agent-os/memory/as its own file using the memory format: frontmatter withname, one-linedescription, andtype(user, feedback, project, or reference), then the fact stated plainly, with Why and How to apply lines for feedback and project types. Add a one-line pointer toMEMORY.md. One memory, one fact, one file. Never save what the work already records or what only matters for this conversation. - Skills: when I describe the same process twice, propose saving it as a skill in
agent-os/methods/(when to use, steps, rules, what done looks like), and add it toTOOLKIT.mdkeyed by intent. - The index: whenever you add a method, tool, or rule, update
TOOLKIT.mdin the same piece of work. A stale index is worse than none. - Hard rules: the rules at the top of
TOOLKIT.mdapply to everything, every session, no exceptions.
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.