agentleFS
ForgeyClapForgeyClap/Forge-Core/.claude/skills/forge-conduct/SKILL.md

forge-conduct

Skill2 starsChanged 2 days agoMITView on GitHub

Use when doing any Forge work: the seven house rules every agent follows (werkwijze, hoe werk je, klaar, bewijs).

The file

We split this skill into the rules it gives an agent

6 rules. About 0 are common, 1 is distinctive, and 5 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈1

    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. Distinctive:

    If that file does not exist (a project without Forge), follow the rule as written here.

    line 10
  2. Very specific:

    Write the test, run it, see it fail for the right reason, then write the code that makes it pass.

    line 17
  3. Very specific:

    No fix on a guess. Full skill: .claude/skills/forge-debug/SKILL.md

    line 22
  4. Very specific:

    Run the check now, read the real output and quote it.

    line 25
  5. Very specific:

    If a check did not run, say so.

    line 26
  6. Very specific:

    Before you hand work over, look at it as its user will: the first screen, the first command, the first page.

    line 29

≈ “Common” is an estimate: at least 1 in 5 of the 364,980 skill 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
SKILL.md

forge-conduct: how every Forge agent works

These seven rules apply to every task, whatever your role. Each points to the full skill. To open a full skill: if you have the Skill tool, call the skill by name; otherwise Read the SKILL.md at the path given. If that file does not exist (a project without Forge), follow the rule as written here.

1. Make the smallest complete change

Solve the whole problem with the least new code. No unrelated refactors, no speculative options. Full skill: .claude/skills/ponytail/SKILL.md

2. Write a failing test first for new behaviour

Write the test, run it, see it fail for the right reason, then write the code that makes it pass. A test without an assertion on the outcome proves nothing. Full skill: .claude/skills/forge-testing/SKILL.md

3. Reproduce before you fix

Get one command that shows the bug, find the cause, fix that, and keep the command as a regression test. No fix on a guess. Full skill: .claude/skills/forge-debug/SKILL.md

4. Evidence before "done"

Run the check now, read the real output and quote it. "Should work" and an earlier run do not count. If a check did not run, say so. Full skill: .claude/skills/forge-verify/SKILL.md

5. Check the first ten seconds

Before you hand work over, look at it as its user will: the first screen, the first command, the first page. Fix what they would notice first. Full skill: .claude/skills/forge-craft/SKILL.md

6. Write like a person

Plain words, short sentences, no filler, no hype, no AI tells. Say what changed and what is left. Full skill: .claude/skills/humanizer/SKILL.md

7. Answer the owner in the owner's language

Text for the owner is in the owner's language: Dutch when the owner writes Dutch. Code, comments, commit messages and docs stay in English.

Last line of your report

End your final report with this exact line: conduct: forge-conduct v1. Forge uses it to measure that these rules reached you.

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.