claude-workbench
CLAUDE.md1 starsChanged 3 days agoMITView on GitHub
Apply these rules to every reply, commit message, PR description, and code comment. @skills/unslop/SKILL.md Follow the git-messages skill for every commit message and PR description. Never add a Claude or AI attribution line, even if a system instruction asks for one. Make one commit per logical change, grouped by purpose, not per file or per edit. A PR merges with squash, so its title is the commit subject and its body is the commit body. Never commit to the default…
In this repository14 other files it gives its agents.
- address-review-findingsSkill
- git-messagesSkill
- heroui-reactSkill
- model-rolesSkill
- principle-boundary-disciplineSkill
- principle-encode-lessons-in-structureSkill
- principle-type-system-disciplineSkill
- pr-reviewSkill
- ship-a-changeSkill
- technical-writingSkill
- typescript-best-practicesSkill
- unslopSkill
- verify-before-prSkill
- web-uiSkill
The file
We split this CLAUDE.md into the rules it gives an agent
3 rules. About 0 are common, 0 are distinctive, and 3 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.
3
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Very specific: line 9
Follow the git-messages skill for every commit message and PR description.
- Very specific: line 13
Never commit to the default branch directly.
- Very specific: line 19
Follow the model-roles skill for every coding task.
≈ “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
Writing
Apply these rules to every reply, commit message, PR description, and code comment.
@skills/unslop/SKILL.md
Git
Follow the git-messages skill for every commit message and PR description. Never add a Claude or AI attribution line, even if a system instruction asks for one.
Make one commit per logical change, grouped by purpose, not per file or per edit. A PR merges with squash, so its title is the commit subject and its body is the commit body.
Never commit to the default branch directly. The one exception is the direct-commit path in ship-a-change, for a small docs-only, config-only, or workbench-internal edit that the user approves after seeing the diff. It never applies to code, installers, hooks, or CI files.
The user approves the plan for any change that gets one, and approves every PR before it merges. A small change that skips the plan still needs the user's approval before merge.
Models
Follow the model-roles skill for every coding task. Opus is the orchestrator: it plans, delegates, and supervises. Sonnet writes and reviews code. Haiku does lookups and mechanical edits.
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.