Skills
skills/jezweb/claude-skills/open-source-contributions/Gen Agent Trust Hub

open-source-contributions

Pass

Audited by Gen Agent Trust Hub on Feb 17, 2026

Risk Level: SAFE
Full Analysis
  • Indirect Prompt Injection (LOW): The pr-prepare agent ingests untrusted documentation files from target repositories to identify project standards, creating a potential surface for indirect prompt injection.
  • Ingestion points: The agent reads CONTRIBUTING.md, .github/CONTRIBUTING.md, and CODE_OF_CONDUCT.md in agents/pr-prepare.md.
  • Boundary markers: Absent; the agent does not use specific delimiters or instructions to ignore embedded commands within these files.
  • Capability inventory: The agent utilizes Bash and Read tools to execute project-defined commands such as npm run lint and npm test as described in agents/pr-prepare.md.
  • Sanitization: Absent; the skill does not perform escaping or validation of the content read from external repository files.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 17, 2026, 04:43 PM