code-prose
Pass
Audited by Gen Agent Trust Hub on Sep 24, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted content (source code comments and identifiers) to apply stylistic rules, which presents a surface for indirect prompt injection.
- Ingestion points: The skill instructions specify reviewing changes in the current branch or user-provided scopes in SKILL.md.
- Boundary markers: No explicit delimiters or boundary markers are defined to isolate the ingested code prose from the agent's instructions.
- Capability inventory: The skill involves file modification and the execution of existing project checks to verify changes.
- Sanitization: No sanitization or filtering of the ingested comments or identifiers is mentioned.
- [SAFE]: The skill is a standard code-refactoring utility. No evidence of malicious intent, such as data exfiltration, credential theft, or remote code execution, was found. The operations are limited to stylistic improvements of code prose.
Audit Metadata