fable-review
Pass
Audited by Gen Agent Trust Hub on Oct 22, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data in the form of local source code, uncommitted diffs, and related test files. This content is passed to an external orchestration tool for review, creating a surface where malicious instructions hidden in the code (e.g., in comments or string literals) could attempt to manipulate the review output or the agent's logic.
- Ingestion points: The skill reads the 'current environment, including uncommitted work' and 'related code and tests' (SKILL.md).
- Boundary markers: The instructions do not specify the use of delimiters, XML tags, or 'ignore instructions' warnings when preparing the review brief containing the untrusted code content.
- Capability inventory: The skill has the capability to read files from the workspace, discover tool IDs, and invoke an external orchestration tool or alternative harnesses.
- Sanitization: There are no explicit instructions to sanitize, filter, or escape the content of the files before they are provided to the reviewer.
Audit Metadata