code-simplifier
Pass
Audited by Gen Agent Trust Hub on Nov 6, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill focuses on refining recently modified code, which represents an ingestion point for untrusted data that could contain malicious instructions.\n
- Ingestion points: Recently modified code sections in the user session (SKILL.md).\n
- Boundary markers: The instructions do not define explicit delimiters or 'ignore' directives for the processed code.\n
- Capability inventory: No tool access, file-system writing, or network operations are specified in the skill configuration.\n
- Sanitization: There are no sanitization or validation steps for the code content before it is processed by the agent.\n- [SAFE]: The skill documentation includes a reference to Anthropic's official code-simplifier agent for its refinement logic and principles.
Audit Metadata