load-repo-context
Pass
Audited by Gen Agent Trust Hub on Sep 26, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill instructs the agent to ingest and process untrusted content from local repository documentation files.
- Ingestion points: Reads the root
./ABOUT.mdand targeted childABOUT.mdfiles (SKILL.md). - Boundary markers: Absent. The instructions do not define delimiters or provide warnings to ignore instructions embedded within the documentation content.
- Capability inventory: File system read access to traverse and read markdown files. No network operations or file write capabilities are defined in the instructions.
- Sanitization: Absent. The skill does not specify any validation, filtering, or escaping of the content read from the
ABOUT.mdfiles before generating an architectural summary.
Audit Metadata