agentleFS
Sign inStart free
dailyherolddailyherold/skills/context-review/SKILL.md

context-review

Skill0 starsChanged 33 days agoMITView on GitHub

Reviews agent context documents — CLAUDE.md, AGENTS.md, opencode steering docs, system prompts, skill files, or any markdown that guides AI agent behavior — against best practices for conciseness, structure, and AI-friendliness. Use this skill whenever the user wants to review, audit, improve, or get feedback on a context doc, steering doc, CLAUDE.md, AGENTS.md, or system prompt. Also trigger when the user asks "how's my context doc", "check this CLAUDE.md", "review this steering doc", "is this well-written for an agent", or any request to evaluate agent-facing instructions.

3 other projects use a version of this.

Adopted by

Most recently taken up by

The file

We split this skill into the rules it gives an agent

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

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈6

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 0

    Very specific

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

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    If the user specified a file, use it.

    line 34
  2. Distinctive:

    Read both reference files before reviewing:

    line 38
  3. Distinctive:

    Read the full content of the file being reviewed.

    line 44
  4. Distinctive:

    If a section is significantly improvable, show a before/after

    line 60
  5. Distinctive:

    Keep feedback direct and actionable.

    line 62
  6. Distinctive:

    Ask if the user wants you to apply any of the changes directly.

    line 66

≈ “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

context-review

Reviews agent context documents against best practices and returns concrete, prioritized feedback.

What counts as an agent context document

Any file that provides instructions, conventions, or reference material to an AI agent:

  • CLAUDE.md / .claude/CLAUDE.md
  • AGENTS.md
  • Opencode steering docs (.md files in .opencode/steering/)
  • Kiro steering docs (.md files in .kiro/steering/)
  • System prompts
  • Skill SKILL.md files
  • Any other markdown file written to guide agent behavior

Step-by-step

Step 1 — Identify the target

If the user specified a file, use it. If not, ask which file to review.

Step 2 — Load references

Read both reference files before reviewing:

  • resources/best-practices.md — principles for writing effective context docs
  • resources/review-checklist.md — detailed checklist to apply

Step 3 — Read the target file

Read the full content of the file being reviewed.

Step 4 — Review and report

Apply the checklist to the file. Structure your feedback as:

What's working

  • Bullet list of strengths (be specific, reference actual content)

Issues — High priority

  • Things that meaningfully hurt agent comprehension or usability

Issues — Lower priority

  • Polish items, minor redundancy, style

Suggested rewrites (optional)

  • If a section is significantly improvable, show a before/after

Keep feedback direct and actionable. Reference specific sections or line content. Don't pad with encouragement.

Step 5 — Offer to apply fixes

Ask if the user wants you to apply any of the changes directly.

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