open-source-contribution

Pass

Audited by Gen Agent Trust Hub on Feb 17, 2026

Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
  • [Prompt Injection] (LOW): The skill possesses a surface for indirect prompt injection because it instructs the agent to read and follow untrusted project documentation like CONTRIBUTING.md and CODE_OF_CONDUCT.md. * Ingestion points: Researching the Project section in SKILL.md. * Boundary markers: Absent. * Capability inventory: Subprocess execution via git, gh, and cargo. * Sanitization: Absent.
  • [Command Execution] (SAFE): The skill includes standard commands for version control and project building (git, gh, cargo). These are legitimate for the stated use case of open-source development and do not include dangerous flags or network redirects to untrusted sources.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 17, 2026, 06:36 PM