What problem does it solve? Creating consistent, high-quality AI agent skills from the Trail of Bits Testing Handbook requires manual analysis of dozens of handbook sections, template selection, cross-reference management, and validation. This meta-skill automates that entire pipeline. ## Core Features & Use Cases - Handbook Discovery and Planning: Scans the Testing Handbook repository, classifies sections by skill type (tool, fuzzer, technique, domain), filters drafts and GUI-only tools, and presents a generation plan for user approval. - Two-Pass Parallel Generation: Spawns parallel agents to generate skill content first, then populates cross-references between skills in a second pass to avoid broken links. - Validation and Self-Improvement: Runs an automated validator for YAML frontmatter, required sections, line limits, and Hugo shortcode removal, then captures lessons to improve templates and discovery logic. - Use Case: Ask the agent to generate skills from the testing handbook, and it will locate the repository, propose a plan covering sections like libFuzzer, Semgrep, and Wycheproof, generate each skill with the correct template, validate them, and update the README cross-reference graph. ## Quick Start Ask the agent to generate skills from the testing handbook and approve the discovery plan it presents.
