What problem does it solve? Determining whether an open-source repository has been compromised—through force-pushed commits, deleted issues, maintainer account takeover, or malicious dependency injection—requires correlating evidence across git history, GitHub APIs, BigQuery archives, and the Wayback Machine, which is slow and error-prone when done manually. ## Core Features & Use Cases - Multi-Source Evidence Collection: Coordinates five specialist investigators covering local git analysis, GitHub REST API, Wayback Machine CDX, GitHub Archive via BigQuery, and IOC enrichment. - Deleted Content Recovery: Recovers force-pushed commits, deleted PRs, and erased issues using direct SHA fetch, git fsck dangling objects, and archived snapshots. - Evidence-Backed Reporting: Manages a SHA-256 integrity-checked evidence store and generates structured forensic reports with validated hypotheses and chain of custody. - Use Case: When a maintainer suspects a supply chain attack (e.g., a suspicious workflow change or a typosquatted dependency), run a full 7-phase investigation to produce a verdict—Compromised, Clean, or Inconclusive—with cited evidence for every claim. ## Quick Start Ask the agent to investigate the GitHub repository owner/repo for signs of supply chain compromise and produce a forensic report.