skills.rest
Maelwalser
MaelwalserCOMMUNITY
Explore all skills by this author →
GitHub

skill-comply

Convert prose skills into behavioral checks and verify agent tool-call traces.

Updated Aug 23, 2026
Search Tags:#tdd#compliance#reporting#agent-testing#stream-json#scenario-generation#llm-classification
One-click install
npx skills add https://github.com/Maelwalser/claude-config --skill skill-comply-maelwalser
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: skill-comply
Source: https://github.com/Maelwalser/claude-config/tree/main/skills/skill-comply
Command: npx skills add https://github.com/Maelwalser/claude-config --skill skill-comply-maelwalser

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires pyyaml, and includes scripts (resource) components.

What problem does it solve?

Detects whether coding agents actually follow documented skills, rules, or agent definitions by converting prose into observable checks and verifying agent behavior against those expectations, eliminating guesswork about whether policies are being applied in practice.

Core Features & Use Cases

  • Auto-generate an expected behavioral sequence (compliance spec) from a skill or rule document and produce structured detectors for each step.
  • Create three scenarios with decreasing prompt strictness, execute agent runs via claude -p capturing stream-json tool-call traces, classify tool calls with an LLM, and enforce deterministic temporal ordering to compute compliance rates.
  • Produce self-contained Markdown reports including specs, prompts, per-scenario compliance scores, timelines, and optional hook-promotion recommendations for low-compliance steps. Use case: validate TDD workflows, verify security or testing rules, and audit agent adherence after rule changes.

Quick Start

Evaluate the skill file at skills/search-first/SKILL.md to generate a compliance spec, run three scenarios, and produce a Markdown report with timelines and hook promotion recommendations.

Frequently Asked Questions about skill-comply

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

FAQPage Schema
How do I measure agent compliance against documented coding rules and skills?▼

Measure agent compliance by converting prose rules into observable behavioral steps, generating scenarios with varying prompt strictness, executing agent runs to capture stream-json tool-call traces, and enforcing deterministic temporal ordering to compute compliance rates.

What is agent compliance testing and how does it verify TDD workflow adherence?▼

Agent compliance testing verifies TDD workflow adherence by extracting YAML specs from rule documents, auto-generating expected behavioral sequences, classifying tool calls with an LLM, and producing Markdown reports with per-scenario compliance scores.

How to generate compliance scenarios from a skill document and execute agent runs?▼

Generate compliance scenarios by pointing to a skill file like SKILL.md to auto-extract a compliance spec, then run three scenarios with decreasing prompt strictness using claude -p to capture stream-json traces for deterministic temporal ordering checks.

Does the compliance reporting tool require specific dependencies or framework setups?▼

The compliance reporting tool requires the pyyaml dependency for YAML spec extraction and operates as scripts components, executing agent runs via claude -p to capture stream-json tool-call traces for LLM-based classification and temporal ordering verification.

What is the best way to audit agent adherence after updating rule documents?▼

Audit agent adherence after rule changes by generating supportive, neutral, and competing prompt scenarios, collecting stream-json tool-call traces from executed runs, and producing self-contained Markdown reports with timelines and hook promotion guidance for low-compliance steps.

Why are my coding agents not following documented rules despite explicit instructions?▼

Agents may fail to follow rules because compliance is not verified behaviorally; detect policy application by converting prose into observable checks, running scenarios at decreasing prompt strictness, and enforcing deterministic temporal ordering against captured tool-call traces.

Related Skills

Explore More Software Engineering Skills→
agentmattersagentmatters
3

skill-comply

Generate compliance specs, scenarios, and reports from skill files.

Official
Advanced
parcadeiparcadei
3.9k

mot

Audit Continuous Claude skills, agents, hooks, and memory for issues.

Community
Intermediate
riftzen-bitriftzen-bit

skill-comply

Generate compliance specs and classify tool-call traces from claude -p stream-json outputs.

Community
Advanced
richardnpaulrichardnpaul

skill-comply

Generate compliance reports by classifying agent tool calls against behavioral specs.

Community
Advanced