guarding-agent-directives

Validate agent directive file changes with a five-question verification workflow.

Updated Nov 9, 2025
One-click install
npx skills add https://github.com/ether-moon/skill-set --skill guarding-agent-directives
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: guarding-agent-directives
Source: https://github.com/ether-moon/skill-set/tree/main/plugins/skill-set/skills/guarding-agent-directives
Command: npx skills add https://github.com/ether-moon/skill-set --skill guarding-agent-directives

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill prevents low-value, duplicated, or unsafe additions to agent directive files by enforcing a structured verification workflow so that only concise, actionable, and project-relevant instructions are added to CLAUDE.md, AGENTS.md, or their referenced documents.

Core Features & Use Cases

  • Five-question verification: Automatically evaluates Recurrence, Non-obviousness, Novelty, Actionability, and Project-specificity with one-line reasoning for each question.
  • Placement recommendation & diffs: Recommends whether content belongs in the core directive, a referenced file, or a skill-internal reference and presents an exact diff for user approval.
  • Review workflow & safety gates: Scans target files for duplicates, contradictions, and superseded items, offers revision options, and requires explicit confirmation before applying changes.
  • Use Case: Use when a user requests adding guidance to CLAUDE.md or when an autonomous agent attempts to modify AGENTS.md to ensure changes are necessary, precise, and safe.

Quick Start

Ask the skill to evaluate a proposed addition to CLAUDE.md, run the five verification questions, and return a pass/fail summary with placement recommendation and an editable diff for confirmation.

Frequently Asked Questions about guarding-agent-directives

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

FAQPage Schema
How do I validate changes to CLAUDE.md or AGENTS.md files?

To validate changes to CLAUDE.md or AGENTS.md files, use a structured five-question verification workflow that checks for recurrence, non-obviousness, novelty, actionability, and project-specificity before applying any edits.

What's the best way to prevent autonomous agents from adding duplicate or unsafe directives?

Preventing duplicate or unsafe agent directives requires a review workflow that scans target files for duplicates, contradictions, and superseded items, offering revision options and requiring explicit user confirmation before applying changes.

How does directive placement recommendation work for agent configuration files?

Directive placement recommendation evaluates proposed content and suggests whether it belongs in the core directive file, a referenced document, or a skill-internal reference, presenting an exact diff for user approval.

When do I need to review agent directive files during repository maintenance?

You need to review agent directive files when a user requests adding guidance to CLAUDE.md or when an autonomous agent attempts to modify AGENTS.md, ensuring changes are necessary, precise, and safe.

Can I require explicit confirmation before an agent modifies AGENTS.md?

Yes, you can require explicit user confirmation before an agent modifies AGENTS.md by enforcing safety gates that present exact diffs and block changes until manual approval is granted.

Why are my agent directive files accumulating low-value or duplicated instructions?

Agent directive files accumulate low-value or duplicated instructions when additions lack a verification workflow enforcing conciseness, actionability, and project-relevance checks before content is committed.