duck-review
Pass
Audited by Gen Agent Trust Hub on Sep 16, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONNO_CODE
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and analyze untrusted external data, specifically code diffs and pull request descriptions, which creates an attack surface for indirect prompt injection.
- Ingestion points: Inputs are explicitly identified in
SKILL.mdas "diff, PR text, or pasted code chunk". - Boundary markers: The instructions do not specify the use of delimiters or "ignore" instructions to separate the untrusted code content from the agent's execution logic.
- Capability inventory: The skill provides instructions for the agent to generate review findings but does not restrict the agent's underlying capabilities (such as tool execution or file system access) while it processes the untrusted data.
- Sanitization: There are no instructions for sanitizing, validating, or escaping the external content before it is processed by the model.
- [NO_CODE]: The skill consists entirely of markdown instructions and reference files. No executable scripts (Python, JavaScript, shell) are included, which reduces the risk of direct command execution or persistence attacks.
Audit Metadata