vicious-mockery
Pass
Audited by Gen Agent Trust Hub on Oct 11, 2026
Risk Level: SAFENO_CODE
Full Analysis
- [SAFE]: The skill contains no source code, shell scripts, or binary executables; it functions purely as a text-based prompt for the agent.
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and analyze user-supplied text (such as plans, designs, or code), creating an indirect prompt injection surface. However, the risk is negligible because the skill does not use any tools or perform any actions (like file system writes, shell access, or network requests) that could be exploited by malicious input.
- [INDIRECT_PROMPT_INJECTION_EVIDENCE]: 1. Ingestion points: User-provided target in SKILL.md Step 2. 2. Boundary markers: Present in Step 1 (requires restating no-touch boundaries). 3. Capability inventory: None identified (text-only output). 4. Sanitization: Absent.
Audit Metadata