Protects sensitive data by scanning AI chat history for secrets and blocking accidental commits.
SpecStory Guard provides a security-focused pre-commit workflow designed specifically for developers using AI coding assistants. It automatically scans the .specstory/history directory for leaked API keys, tokens, and private credentials before they reach your repository. By integrating directly into the Git lifecycle, it ensures that conversational artifacts—which often contain sensitive environment variables or debugging tokens—are redacted or removed, preventing costly security breaches and maintaining repository hygiene.
Key Features
01Advanced pattern matching for AWS keys, GitHub tokens, and private keys
02Automated pre-commit git hook integration
03Customizable allowlist for false positives via regex
04Manual on-demand scanning of history files
052 GitHub stars
06Detailed reporting with redacted previews for safe remediation
Use Cases
01Auditing existing AI-generated history files for sensitive credentials
02Enforcing security guardrails across team repositories using AI coding tools
03Preventing accidental leakage of API keys shared during AI chat sessions
