how
Pass
Audited by Gen Agent Trust Hub on May 16, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill processes external, potentially untrusted data from codebase files, creating a surface for indirect prompt injection.\n
- Ingestion points: The skill ingests codebase data using
rgand direct file reading as specified in theWorkflowsection ofSKILL.mdand the instructions inreferences/explorer-prompt.md.\n - Boundary markers: There are no explicit delimiters or instructions provided in the synthesis or exploration templates to differentiate between the agent's instructions and the content of the files being read, nor are there instructions to ignore instructions found within the code.\n
- Capability inventory: Across
SKILL.mdand the referenced templates, the skill demonstrates capabilities for file system exploration, reading file contents, and orchestrating subagents to process different slices of the data.\n - Sanitization: No sanitization or validation of the ingested code content is performed before the data is synthesized into the final output.
Audit Metadata