fable-review
Pass
Audited by Gen Agent Trust Hub on Oct 7, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: No malicious patterns or security vulnerabilities were detected in the skill instructions, metadata, or supporting documentation.
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted input in the form of implementation_diff data. 1. Ingestion points: Untrusted data enters the agent context through the implementation_diff input parameter defined in SKILL.md. 2. Boundary markers: The skill lacks explicit markers or instructions to treat data within the diff as potentially untrusted instructions. 3. Capability inventory: The skill has no associated scripts and is restricted from mutating the workspace (mutatesWorkspace: false), significantly limiting the potential impact of an injection. 4. Sanitization: No specific sanitization or filtering logic is present to pre-process the diff content. This risk is mitigated by core instructions requiring skepticism and evidence-grounded findings.
Audit Metadata