codebase-onboarding
Pass
Audited by Gen Agent Trust Hub on Sep 10, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it reads and processes untrusted external data (the contents of a repository) and uses it to generate output.
- Ingestion points: The skill instructs agents to explore and read the entire codebase, including directory structures, schemas, API routes, and deployment configurations (SKILL.md, lines 12-32).
- Boundary markers: No specific delimiters or safety instructions are used to distinguish between the repository's data and the agent's instructions during the synthesis phase.
- Capability inventory: The skill has the capability to write files to the local project root, specifically creating or overwriting
ONBOARDING.md(SKILL.md, line 54). - Sanitization: There is no evidence of sanitization or validation of the content extracted from the repository before it is written to the output file.
Audit Metadata