implement-small-change
Pass
Audited by Gen Agent Trust Hub on Oct 23, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to read and process local repository data, which creates a surface for indirect prompt injection if the source files or instructions contain malicious content.
- Ingestion points: Accesses repository source code, git diffs, and local documentation files as described in
SKILL.md. - Boundary markers: The instructions emphasize preserving user changes and establishing boundaries, but they do not provide specific technical delimiters for data interpolation.
- Capability inventory: The skill allows the agent to execute local validation commands such as linters, test runners, and build tools.
- Sanitization: There are no explicit instructions for escaping or sanitizing repository-sourced content before it is processed or used in commands.
Audit Metadata