agentleFS
Sign inStart free
willey-labswilley-labs/agent-skills/skills/writing-standards/SKILL.md

writing-standards

Skill3 starsChanged 33 days agoMITView on GitHub

Standards for writing, editing, and reviewing DOCUMENTS — READMEs, specs, rules, skills, design docs, guides, and any prose deliverable produced from a source (existing code, or a prior discussion). Two rule sets apply on every document task: (1) source-to-deliverable — code→doc describes what the system DOES with no code or code-identifiers; discussion→rule states the general principle, not the specific example just discussed; (2) anti-slop — cut hedging, hype words, throat-clearing, cheerleading, reflexive formatting, and padding. Use when the user says "write a document", "document this code", "write up what this does", "turn this into a rule/skill", "make a doc from our discussion", "review this doc", or "is this slop?". Consult before authoring or reviewing ANY document, even when the user does not say "standards" — every document write/edit/review must comply. The companion to coding-standards: that one governs code, this one governs documents.

1 other project uses a version of this.

Adopted by

Most recently taken up by

  • Other versions1 project uses a version of this.

    Word for word · 0

    None found.

    Edited · 0

    None found.

    Wrote their own version · 1

    1. tsukhani/jclaw★ 42same name

    How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.

  • In this repository3 other files it gives its agents.

The file

We split this skill into the rules it gives an agent

4 rules. About 0 are common, 3 are distinctive, and 1 is very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈3

    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
  1. Distinctive:

    Every document you write, edit, or review must comply with these rules.

    line 23
  2. Very specific:

    All paths in this document are relative to this SKILL.md file, so they resolve wherever the skill is installed.

    line 34
  3. Distinctive:

    Read both common references, then produce the document.

    line 74
  4. Distinctive:

    When reviewing a document, report violations by rule code with the offending line and the fix — the same way coding-standards reports a …

    line 83

≈ “Common” is an estimate: at least 1 in 5 of the 350,495 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

Writing Standards

Every document you write, edit, or review must comply with these rules. A document is any prose deliverable: a README, a spec, a rule, a skill, a design doc, a guide, a review, release notes. This skill does not govern normal conversation — only the deliverable.

The rules come in two sets, both applied on every document task:

  • references/common/source-to-deliverable.md — how to turn a source (code you read, or a discussion you had) into a deliverable without dragging the source into the output.
  • references/common/anti-slop.md — the patterns that make a document read as machine-padded filler, each with a rule code and a fix.

All paths in this document are relative to this SKILL.md file, so they resolve wherever the skill is installed.

Step 0 — Bootstrap the reminder hooks (once per session, only if needed)

The hooks inject a short reminder of these rules at session start and on every prompt, so the rules don't get buried as a conversation grows. Run the fast read-only check first (single absolute-path command, no cd/&&):

python3 <skill-dir>/bootstrap.py --verify

Exit 0 → hooks are wired; go to Step 1. Non-zero → ask the user before wiring them. Say what it changes: two entries — SessionStart and UserPromptSubmit — added to this scope's settings file (this project's .claude/settings.json, or ~/.claude/settings.json for a global install), each printing a short reminder; existing hooks are kept and nothing is installed. On a yes:

python3 <skill-dir>/bootstrap.py

Then tell the user to restart the session so the hooks activate; until then the reminder isn't injected. On a no, don't ask again this session. Either way this SKILL.md is in force for any document you write this session.

Step 1 — Pick the document type (optional refinement)

references/common/ applies to every document. If a per-type guide exists under references/<doc-type>/ (README, spec, review, …), read it too — it specializes the common rules for that kind of document. If none matches, the common rules are the whole standard.

Step 2 — Write, edit, or review against the rules

Read both common references, then produce the document. The two failures this skill exists to stop:

  1. Echoing the source. You read code (or had a discussion) to understand something, then paste that source — the code, the function names, the one example you just discussed — into the deliverable. The source is what you learned from; it does not appear in the output. See source-to-deliverable.md.
  2. Slop. Hedging, hype, throat-clearing, cheerleading, reflexive headers/bold/lists, and sentences that sound like content but carry none. See anti-slop.md.

When reviewing a document, report violations by rule code with the offending line and the fix — the same way coding-standards reports a code review.

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