4-step-program

Pass

Audited by Gen Agent Trust Hub on Apr 25, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill instructs the agent to execute GitHub CLI commands such as gh issue view and gh pr view to manage tasks and verify pull request status.
  • [PROMPT_INJECTION]: The skill facilitates indirect prompt injection by ingesting untrusted data from external sources.
  • Ingestion points: GitHub issue bodies are read via gh issue view in SKILL.md (Step 1 and Step 3.5).
  • Boundary markers: No explicit delimiters or instructions to ignore embedded commands within the issue body are provided.
  • Capability inventory: The agent can call assign_task, send_message_to_agent, and execute shell commands through the GitHub CLI.
  • Sanitization: There is no evidence of sanitization, validation, or escaping of the content retrieved from GitHub issues before it is used to generate delegation prompts for other agents.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 25, 2026, 06:51 PM
Security Audit — agent-trust-hub — 4-step-program