lf-new-feature
Pass
Audited by Gen Agent Trust Hub on Apr 6, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface (Category 8) by design. It reads and processes content from discovery.md and other input files to populate technical briefing templates.\n- Ingestion points: Files located in ai/specs//discovery.md and inputs/ are read and their contents are used to drive the generation of new documentation.\n- Boundary markers: There are no explicit instructions to use delimiters or ignore instructions within the processed files, increasing the risk that embedded commands in the inputs could be followed by the agent.\n- Capability inventory: Uses Glob and Read tools to access local project files. The output is a Markdown documentation file (briefing-tech.vN.md).\n- Sanitization: No validation or sanitization of the input text is performed before it is included in the generated document or processed as part of the technical lead logic.
Audit Metadata