prps-agentic-engineering

Pass

Audited by Gen Agent Trust Hub on May 17, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill's installation process requires cloning a remote repository from a third-party GitHub account (https://github.com/Wirasm/PRPs-agentic-eng.git) and copying executable scripts into the local environment.
  • [COMMAND_EXECUTION]: The methodology involves creating and executing shell scripts (e.g., '.claude/hooks/prp-ralph-stop.sh') and requires granting execution permissions via 'chmod +x'.
  • [COMMAND_EXECUTION]: The 'Ralph Loop' functionality relies on the autonomous execution of project-specific build and test commands such as 'npm test', 'npm run build', and 'npm run lint' to validate AI-generated code changes.
  • [PROMPT_INJECTION]: The skill implements an indirect prompt injection surface through the '/prp-issue-investigate' command.
  • Ingestion points: Data is fetched from external GitHub issues (attacker-controlled content).
  • Boundary markers: None identified in the provided instructions to differentiate issue content from agent instructions.
  • Capability inventory: The skill has capabilities for file system modification, git operations, and subprocess execution across multiple scripts.
  • Sanitization: There is no evidence of sanitization or filtering of the content retrieved from GitHub before it is processed by the agent to create implementation plans.
Audit Metadata
Risk Level
SAFE
Analyzed
May 17, 2026, 09:25 PM
Security Audit — agent-trust-hub — prps-agentic-engineering