agentleFS
Sign inStart free
HypercubedHypercubed/Agent-Knowledge-Starter-Kit/.agents/AGENTS.md

Agent-Knowledge-Starter-Kit

AGENTS.md7 starsChanged 2 months agoMITView on GitHub

This repository uses the AKSK knowledge loop for durable learning:

  • Where files are headingNew AGENTS.md files have more rules about keeping secrets out of the code, and fewer about code style.

    Of the AGENTS.md files we read, the 31,439 first seen in the last six months, against the older ones: the kind of rule that gained most share, and the kind that lost most. About files like this in general, not this one, and an estimate, since a kind is a pattern in the words.

  • In this repository8 other files it gives its agents.

The file

We split this AGENTS.md into the rules it gives an agent

5 rules. About 0 are common, 0 are distinctive, and 5 are 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.

  • 5

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Very specific:

    WHEN EXPLORING KNOWLEDGE: Start from openwiki/index.md and the directory indexes under openwiki/decisions/ and openwiki/troubleshooting/.

    line 4
  2. Very specific:

    WHEN DEBUGGING: If you encounter a failing test, build error, or runtime exception, your FIRST action MUST be to search durable knowledge …

    line 5
  3. Very specific:

    Skill renaming workflow: When renaming a skill, use shell tools to locate all path and string references and update the SKILL.md …

    line 10
  4. Very specific:

    Maintenance script feedback: When writing scripts (especially those using npx), always include an initial "Starting..." message and a …

    line 11
  5. Very specific:

    OpenWiki concurrency guard: Before editing files when openwiki may be active, check test -f openwiki/.run.json && echo "openwiki running: …

    line 12

≈ “Common” is an estimate: at least 1 in 5 of the 40,246 AGENTS.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.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
AGENTS.md

Routing Directives

  • UPON STARTUP: You MUST read openwiki/index.md and .agents/AGENTS.md before executing any file modifications. This ensures you understand the repository layout and available tools.
  • WHEN EXPLORING KNOWLEDGE: Start from openwiki/index.md and the directory indexes under openwiki/decisions/ and openwiki/troubleshooting/. Indexes contain descriptions and tags that help identify relevant files.
  • WHEN DEBUGGING: If you encounter a failing test, build error, or runtime exception, your FIRST action MUST be to search durable knowledge for it: grep -ri "<error or symptom>" openwiki/ .agents/ before debugging blind.
  • BEFORE ARCHITECTURAL CHANGES: You MUST search openwiki/decisions/ for recorded decisions to ensure your proposed changes do not violate established design patterns.

Project Learnings

  • Skill renaming workflow: When renaming a skill, use shell tools to locate all path and string references and update the SKILL.md frontmatter. Add a note under "Recurring pitfalls" in AGENTS.md for this pattern.
  • Maintenance script feedback: When writing scripts (especially those using npx), always include an initial "Starting..." message and a --verbose flag to prevent agents from assuming a hang during background execution.
  • OpenWiki concurrency guard: Before editing files when openwiki may be active, check test -f openwiki/.run.json && echo "openwiki running: $(jq -r .phase openwiki/.run.json)"; if running, wait or notify the user — editing mid-run leaves openwiki/.last-update.json: status: "interrupted" and forces the next run to re-plan (source fingerprint drift).

Self-improvement loop

This repository uses the AKSK knowledge loop for durable learning:

  1. Analyze — was a rule missing from .agents/AGENTS.md or ignored?
  2. Closeout — bundle the session under .agents/sessions/ per task-closeout.
  3. Distill — promote stable lessons: decisions/troubleshooting to openwiki/, behavior rules to .agents/AGENTS.md or playbooks.
  4. Prune — remove guidance that no longer prevents mistakes.

Download now (ZIP)

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
What did you use it for? e.g. Reviewing pull requests in a Rails app

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.

Check your setup, free