lore-check
Pass
Audited by Gen Agent Trust Hub on Oct 1, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes data from several ingestion points which could contain malicious instructions designed to influence the agent.
- Ingestion points: The skill reads files from the
docs/lore/directory,README.md,CONTEXT.md, and a feedback log file.lore-feedback.jsonl. - Boundary markers: There are no specified boundary markers or instructions for the agent to ignore potential commands embedded within the processed documentation or log data.
- Capability inventory: The skill utilizes file system tools for reading, grepping, and writing (specifically updating a heartbeat timestamp in
.lore-last-check). - Sanitization: No sanitization or validation routines are mentioned for the content being audited.
- [COMMAND_EXECUTION]: The skill involves file system interactions, including using 'Read' and 'Grep' tools directly and performing a specific write operation to
docs/lore/.lore-last-check. While these are functional requirements for a documentation audit tool, they represent the execution of file-system operations based on the skill's instructions.
Audit Metadata