grimoire-slice
Pass
Audited by Gen Agent Trust Hub on Nov 11, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: No security issues were detected. The skill focuses on documentation and task decomposition with clear boundaries on its capabilities.
- [INDIRECT_PROMPT_INJECTION]: The skill possesses an input surface for external specifications and conversation history. The risk is assessed as safe given the restricted output format and lack of dangerous capabilities. 1. Ingestion points: Processes "current spec" and "conversation requirements" as described in SKILL.md. 2. Boundary markers: The skill is instructed to follow supplied knowledge boundaries and permissions; however, it lacks explicit syntax for ignoring embedded instructions in data. 3. Capability inventory: File writing is limited to markdown tickets within the .grimoire/ticket/ directory; there are no network operations, subprocess executions, or credential access patterns. 4. Sanitization: No specific sanitization or escaping of external content is implemented, but the lack of an execution context for the generated markdown files mitigates the potential impact of injected instructions.
Audit Metadata