github-readme

Pass

Audited by Gen Agent Trust Hub on Sep 17, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill is designed with a strong security posture, including mandatory PII and infrastructure scrubbing to prevent the accidental publication of sensitive information in repository documentation.
  • [COMMAND_EXECUTION]: Uses standard, non-privileged development tools such as git, ls, and gh (GitHub CLI) to extract repository metadata. These operations are limited to gathering technical context (e.g., tech stack detection, remote URLs) and do not pose a risk of system compromise.
  • [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data from local repository files (e.g., package.json descriptions, source code comments) to generate summaries. While this creates a potential surface for indirect injection, the risk is mitigated by the skill's prescriptive output templates and its built-in security auditing features.
  • Ingestion points: Reads local project files including package.json, pyproject.toml, Cargo.toml, and existing README.md files.
  • Boundary markers: The skill uses predefined markdown templates to structure the output rather than direct interpolation of raw file contents.
  • Capability inventory: Local filesystem read/write access and execution of standard version control CLI tools.
  • Sanitization: Implements an extensive "PII / Infrastructure Scrub" using regex to detect and remove AWS keys, GitHub tokens, Slack webhooks, private IP addresses, and other sensitive identifiers.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 17, 2026, 08:54 PM
Security Audit — agent-trust-hub — github-readme