What problem does it solve? AI coding sessions often capture sensitive data like API keys, tokens, and private keys in chat history files. This Skill prevents those secrets from being accidentally committed to your git repository by scanning .specstory/history before every commit. ## Core Features & Use Cases - Pre-Commit Hook Installation: Installs a git pre-commit hook that automatically scans history files on every commit and blocks the commit if secrets are found. - Secret Pattern Detection: Detects AWS keys, GitHub/GitLab/Slack tokens, private keys, JWTs, Stripe keys, and generic credential assignments using regex patterns. - Allowlist Tuning: Supports custom allowlist regex patterns via the SPECSTORY_GUARD_ALLOWLIST environment variable to suppress false positives like placeholders and examples. - Use Case: After pasting an API key into an AI chat to debug an integration, run a manual scan to find the leaked key in your history files, redact it, and rotate the credential before committing. ## Quick Start Ask the AI to install the specstory-guard pre-commit hook and run a scan of your .specstory/history directory for potential secrets.