Go web crawler to scrape documentation sites and convert content to clean Markdown for LLM ingestion (RAG, training data).

Scan any agent skill — authored or compiled — for prompt injection, secrets, and malicious execution before it touches your agent. Or compile the long tail on the fly.