explain-deeply
Pass
Audited by Gen Agent Trust Hub on Oct 27, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest data from untrusted or external sources to provide explanations, creating a surface for indirect prompt injection.
- Ingestion points: Repo files (local codebase) and external information (via fetching) are explicitly used as the source of truth in Step 1.
- Boundary markers: There are no specific instructions for the agent to use delimiters or to ignore embedded instructions in the source files or fetched content.
- Capability inventory: The skill implies the use of file system read operations and network fetch operations to gather information.
- Sanitization: No sanitization or validation of the content read from files or external sources is specified.
Audit Metadata