skills/paulund/ai/plan-bug/Gen Agent Trust Hub

plan-bug

Pass

Audited by Gen Agent Trust Hub on May 9, 2026

Risk Level: SAFE
Full Analysis
  • [COMMAND_EXECUTION]: The skill executes the GitHub CLI (gh issue create) to create issues on the remote repository. This operation is the core functionality of the skill and is triggered only upon user confirmation.
  • [PROMPT_INJECTION]: The skill processes user-supplied data (bug descriptions, reproduction steps) and interpolates them into a shell command. It effectively mitigates indirect prompt injection risks by:
  • Confirmation Gate: Forcing the agent to display the full draft and wait for explicit user approval before execution.
  • Quoted Heredocs: Using the cat <<'EOF' shell pattern, which ensures the issue body is treated as a literal string and prevents the shell from executing any command substitutions or variable expansions contained within the user input.
  • [DATA_EXFILTRATION]: Issue data is sent to GitHub repositories identified via git remote. This is a standard operation for developer tools and involves a well-known service, presenting no unexpected data exposure risks.
Audit Metadata
Risk Level
SAFE
Analyzed
May 9, 2026, 05:44 PM
Security Audit — agent-trust-hub — plan-bug