skill-gen
Pass
Audited by Gen Agent Trust Hub on Feb 20, 2026
Risk Level: SAFE
Full Analysis
- Indirect Prompt Injection (SAFE): The skill ingests user input to generate instructions. Ingestion points: User requirements provided during the generation workflow in SKILL.md. Boundary markers: Utilizes Markdown structure and YAML frontmatter as delimiters. Capability inventory: Creating directories and writing files to ~/.claude/skills/. Sanitization: Relies on structured templates to guide output. Severity is SAFE as this is the primary functionality.
- Command Execution (SAFE): The workflow includes creating local directories and running validation scripts. These are standard developer operations and do not include any obfuscated or remote commands.
- Dynamic Execution (SAFE): The skill guides the creation of local scripts from templates. No evidence of untrusted remote script execution or unsafe deserialization methods was found.
Audit Metadata