skills.rest
siarhei-belavus
siarhei-belavusCOMMUNITY
Explore all skills by this author →
GitHub

review-agents-tree

Audits the tracked AGENTS.md tree and writes a review report with findings and proposals.

2|Updated Apr 29, 2026
Search Tags:#code review#context management#audit#agents.md#conflict detection#repository maintenance#documentation hygiene
One-click install
npx skills add https://github.com/siarhei-belavus/agent-public --skill review-agents-tree-siarhei-belavus
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: review-agents-tree
Source: https://github.com/siarhei-belavus/agent-public/tree/main/skills/atelier-workflow/review-agents-tree
Command: npx skills add https://github.com/siarhei-belavus/agent-public --skill review-agents-tree-siarhei-belavus

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Over time, a repository's tracked AGENTS.md files accumulate bloat, stale references, duplicated guidance, and conflicting rules that degrade agent behavior. This Skill performs a periodic hygiene audit of the entire AGENTS.md tree and produces a structured review report. ## Core Features & Use Cases - Tree-wide audit: Enumerates all tracked AGENTS.md files and checks shape, lean-ness, specificity, hierarchy quality, conflicts, staleness, and task-state leakage. - Safe auto-fixes: Applies only trivial non-semantic hygiene fixes such as heading normalization, whitespace cleanup, and exact-duplicate deduping. - Structured report: Writes .agents/AGENTS_TREE_REVIEW.md with a status, AT# findings ordered by severity, and approval-gated proposals for compaction or sharding. - Use Case: After months of edits, run this audit to detect a bloated 300-line AGENTS.md, a parent/child conflict over test commands, and stale path references, then get a concrete cleanup plan. ## Quick Start Audit the tracked AGENTS.md tree in this repository and write the review report to .agents/AGENTS_TREE_REVIEW.md.

Frequently Asked Questions about review-agents-tree

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I audit AGENTS.md files for conflicts and bloat?▼

Run this audit to enumerate every tracked AGENTS.md file, check each for canonical sections, lean content, specificity, and hierarchy quality, then receive a report with severity-ordered findings and concrete cleanup proposals.

What does the AGENTS tree review report contain?▼

The report at .agents/AGENTS_TREE_REVIEW.md includes a health status, a list of auto-applied hygiene fixes, AT# findings ordered by severity, approval-gated proposals, and a suggested next pass such as compaction or sharding.

Can the audit modify AGENTS.md files automatically?▼

It auto-applies only trivial non-semantic fixes like heading normalization, whitespace cleanup, and exact-duplicate removal. Semantic changes, shard creation, and conflict resolution always require explicit human approval.

When should I run an AGENTS.md tree review?▼

Run it periodically as durable-context hygiene, not as a workflow gate. It is most useful after heavy editing periods when files may have grown past roughly 200 lines, drifted from canonical sections, or accumulated stale paths and commands.

Does the audit treat AGENTS.override.md as canonical context?▼

No. Tracked AGENTS.override.md files are flagged as hygiene smells and checked only for misuse or drift. They are never treated as canonical repository truth during the audit.

Related Skills

Explore More Software Engineering Skills→
d3mingd3ming

claude-agents-md

Audit repository trees for AGENTS.md and CLAUDE.md wrapper configurations.

Community
Intermediate
otahontasotahontas
1

agents-md-improver

Analyze AGENTS.md files and propose quality improvements across repositories.

Community
Basic
PostHogPostHog
713

context-layer-dreaming

Synthesizes recent organizational activity into a durable, sourced context wiki.

Official
Advanced
facebookfacebook
247.1k

fix

Run yarn prettier and yarn linc to fix linting and formatting issues.

Official
Intermediate