curated-skill-creator
Pass
Audited by Gen Agent Trust Hub on Mar 17, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: No malicious instructions or security threats detected. The skill functions as an educational and development tool.
- [EXTERNAL_DOWNLOADS]: Uses WebFetch to retrieve skill templates and documentation from public repositories such as GitHub.
- [COMMAND_EXECUTION]: Assists in creating a directory structure and local files (SKILL.md, scripts) to implement new agent capabilities.
- [PROMPT_INJECTION]: Exposed to potential indirect prompt injection (Category 8) when analyzing external websites. Ingestion points: Web content fetched during the research phase. Boundary markers: None explicitly requested for the agent's internal processing. Capability inventory: Ability to create and write to local files. Sanitization: Relies on a collaborative process where the user must approve and curate the integrated findings.
Audit Metadata