request-refactor-plan

Pass

Audited by Gen Agent Trust Hub on Jun 25, 2026

Risk Level: SAFENO_CODEPROMPT_INJECTION
Full Analysis
  • [SAFE]: The skill implements a standard productivity workflow for planning code refactors through interactive user interviews and repository analysis.
  • [NO_CODE]: The skill consists exclusively of markdown instructions and does not include any scripts or executable files.
  • [PROMPT_INJECTION]: The skill has a surface for indirect prompt injection due to its ingestion of external data.
  • Ingestion points: Local repository content (SKILL.md, steps 2 and 6) and user-provided descriptions (steps 1, 3, 4, and 5).
  • Boundary markers: Absent.
  • Capability inventory: Reading the filesystem and creating GitHub issues.
  • Sanitization: The skill explicitly instructs the agent to avoid including specific file paths or code snippets in the final GitHub issue, providing a mitigation against accidental data exposure or malicious injection playback.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 25, 2026, 12:34 AM
Security Audit — agent-trust-hub — request-refactor-plan