knowledge
Pass
Audited by Gen Agent Trust Hub on Sep 30, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill handles project knowledge records, which involves reading and writing to Markdown files and processing command output. This creates a standard surface for indirect prompt injection common to documentation tools.
- Ingestion points: Markdown prose files (*.md) and the output of the 'doctrine knowledge show' command.
- Boundary markers: No specific boundary markers or 'ignore' instructions are used for the record content.
- Capability inventory: The skill uses a local CLI tool ('doctrine') for managing project records.
- Sanitization: No explicit sanitization of the record text is described.
- [COMMAND_EXECUTION]: The skill instructions rely on the 'doctrine' CLI tool for all functionality. These commands are project-local and do not include network requests, system modifications, or privilege escalation.
Audit Metadata