duck-tidy
Pass
Audited by Gen Agent Trust Hub on Aug 30, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill acts as an audit tool for repository comments and documentation, which are external, untrusted sources that can influence the agent's behavior.
- Ingestion points: In
SKILL.md(Step 1), the skill defines its scope to include code comments, doc comments, and non-CONTEXT markdown files from the worktree. - Boundary markers: While the skill establishes philosophy guardrails and evidence-backed staleness rules (Step 2), it lacks explicit prompt delimiters or 'ignore' instructions for the content of the comments it processes, allowing embedded text to potentially be interpreted as instructions.
- Capability inventory: The skill possesses the capability to execute shell-level git commands (
git diff,git symbolic-ref,git status) and delegate workspace-mutating actions (file edits, deletes, and git commits) through a handoff to theduck-patchutility. - Sanitization: No explicit sanitization, escaping, or schema validation is mentioned for the raw comment text before it is evaluated by the agent.
Audit Metadata