grimoire-tidy
Pass
Audited by Gen Agent Trust Hub on Nov 11, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes the content of user-provided documentation files, creating a potential surface for instructions embedded within that data.\n
- Ingestion points: The skill reads every document proposed for change in full, including related specs, tickets, and evidence files within the .grimoire directory.\n
- Boundary markers: There are no instructions to use delimiters or specific boundary markers to isolate document content from the agent's core instructions.\n
- Capability inventory: The skill possesses the ability to read, modify, and delete files, and to execute Git commands.\n
- Sanitization: No sanitization, escaping, or validation logic is specified for the content read from external files.\n- [COMMAND_EXECUTION]: The skill instructs the agent to execute shell commands for repository inventory.\n
- Evidence: The instructions explicitly require the use of
git ls-filesto inventory every tracked file in the selected root.
Audit Metadata