codebase-context-scan

Pass

Audited by Gen Agent Trust Hub on Jun 22, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection. It ingests content from repository files like CHANGELOG.md and firestore.rules and interpolates them into a summary file intended for consumption by other AI agents.
  • Ingestion points: The script scripts/generate-overview.mjs reads CHANGELOG.md, firestore.rules, and package.json files.
  • Boundary markers: The generated output file lacks explicit delimiters or instructions to ignore embedded content.
  • Capability inventory: The script performs file system reads and writes (fs.readFile, fs.writeFile, fs.readdir) and SKILL.md uses the rg tool.
  • Sanitization: Content is parsed via regular expressions but is not escaped or sanitized against potential instruction injection before being written to the output markdown.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 22, 2026, 01:04 AM
Security Audit — agent-trust-hub — codebase-context-scan