agentsclimarketplace
Sign in

Validate agents md

Skill MSiccDev/ai-context-kit/skills/validate-agents-md

Validate AGENTS.md files for structural completeness, operational contract coverage, neutrality, and repository alignment.From its SKILL.md

Install
npx -y skills add MSiccDev/ai-context-kit --skill validate-agents-md

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

One thing to look at

  • 12 stars12 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.

SKILL.md

2.5 KB, 468 tokens by cl100k_base, as published. Nobody here has run it

Validate AGENTS.md

Purpose

Validate AGENTS.md files to ensure they are concise, operationally complete, and aligned with repository structure and governance.

When To Use

  • Use this skill when adding or updating root or nested AGENTS.md files.
  • Use this skill before accepting AGENTS policy changes.
  • Do not use this skill for user/project instruction validation.

How to Invoke

Load or attach this file's contents into your AI session to activate the workflow (paste, upload, or reference with #file:skills/validate-agents-md/SKILL.md in VS Code Copilot Chat). In Claude Projects, add it to project knowledge. See Invoking Skills in the README for full platform guidance.

Required Inputs

  • Target AGENTS.md path.
  • Repository path context for link/path validation.
  • Policy expectations for session-state and command namespace.

Workflow

  1. Run validation phases from references/phase-checks.md.
  2. Check for <!-- spec_version: ... --> comment: flag as WARNING if absent (pre-existing files may predate this requirement); flag as WARNING if present but older than 1.4.3.
  3. Generate report using references/report-contract.md.
  4. Apply deterministic scoring from references/scoring.md.
  5. Classify findings into critical/warning/enhancement buckets.
  6. Provide concrete fixes for operational contract gaps.

Output Expectations

  • Structured markdown report with five validation phases.
  • Compliance score and pass band.
  • Explicit list of critical issues, warnings, and enhancements.
  • Example fixes for common structural and policy failures.
  • Report schema follows references/report-contract.md.
  • Scoring and grade bands follow references/scoring.md.

Resources

  • Phase checks: references/phase-checks.md
  • Report contract: references/report-contract.md
  • Scoring model: references/scoring.md
  • Example target: ../../AGENTS.md

Constraints And Safety

  • Keep the scoring rubric deterministic; findings should be repository-specific and actionable.
  • Avoid over-prescriptive style-only feedback.
  • Preserve provider-neutral language.
  • Do not duplicate full normative specs when suggesting fixes.

What ships with it: 4 files

3.9 KB alongside SKILL.md

references/

Gives 0 of the 12 instructions most memory context skills give in 468 tokens

Counted across 754 of the 1,056 authors here whose files we hold, read 2026-09-06

  • Preserve existing content structurein 15 of 754, across 9 files
  • Front-load the leading wordin 14 of 754, across 10 files
  • Update existing entries instead of duplicatingin 14 of 754, across 7 files
  • Keep CLAUDE.md under one hundred linesin 14 of 754, across 12 files
  • Read CLAUDE.md at the project rootin 14 of 754
  • Keep each meaning in a single source of truthin 12 of 754, across 8 files
  • Redact sensitive information before committingin 11 of 754, across 4 files
  • Scan for all CLAUDE.md filesin 11 of 754, across 7 files
  • Use frontmatter for metadata on filesin 10 of 754, across 3 files
  • Repeat user interactions 10 timesin 10 of 754, across 4 files
  • Write the CLAUDE.md file into the target folderin 10 of 754, across 8 files
  • Use memlab to process snapshotsin 9 of 754, across 3 files

Said here and by no other author read

  • Run validation phases from phase checks
  • Check for spec version comment
  • Generate report using report contract
  • Apply deterministic scoring
  • Classify findings into buckets
  • Provide concrete fixes for gaps

Grouped from the skills themselves: near-identical wordings counted once, and counted by distinct author, so one author publishing three of these counts once. Length counted with cl100k_base; the agent that loads this file may tokenize it differently.

Keep looking

Skills are one crate of 325,949. Ordering is by how many stacks a row turns up in, so the top of any crate is what has actually been picked rather than what has the most stars.