onboard-codebase
Pass
Audited by Gen Agent Trust Hub on Oct 14, 2026
Risk Level: SAFEDATA_EXFILTRATIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and analyze an entire unmapped codebase. Maliciously crafted comments or strings within the source files could potentially influence the agent's logic or the content of the generated documentation.
- Ingestion points: Project codebase (SKILL.md).
- Boundary markers: None identified to delimit untrusted code from instructions.
- Capability inventory: Extensive file reading across the project directory; file writing to
llmwiki/andAGENT-business.md. - Sanitization: No specific sanitization or filtering of codebase content is mentioned before it is processed or used to generate wiki entries.
- [DATA_EXFILTRATION]: The skill requires access to the full project codebase to perform its analysis. This capability could lead to the exposure of sensitive information (such as environment variables, configuration secrets, or private keys) if they are accidentally included in the generated wiki documentation. While no external network exfiltration is explicitly requested, the transformation of private code data into structured wiki files constitutes a data exposure risk.
Audit Metadata