clean-writing

Pass

Audited by Gen Agent Trust Hub on Sep 11, 2026

Risk Level: SAFENO_CODEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [NO_CODE]: The skill is composed strictly of markdown instructions for prose improvement. It does not specify any allowed-tools in the frontmatter nor does it contain accompanying script files.\n- [SAFE]: The URLs included in the documentation point to legitimate educational and open-source resources (Project Gutenberg, GitHub, and Pearson) and are provided solely for contextual reference.\n- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted user-supplied text for prose editing. 1. Ingestion points: Text passed to the agent for composition or tightening (README.md). 2. Boundary markers: Absent. 3. Capability inventory: None; no tools, network operations, or file system access are present (SKILL.md). 4. Sanitization: Absent. The lack of capabilities results in a safe assessment.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 11, 2026, 06:41 PM
Security Audit — agent-trust-hub — clean-writing