skill-creator
Pass
Audited by Gen Agent Trust Hub on Oct 13, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: Fetches skill creation best practices and workflow patterns from Anthropic's official GitHub repository.
- [INDIRECT_PROMPT_INJECTION]: The skill retrieves external documentation to guide the agent's behavior during skill creation.
- Ingestion points: Retrieves markdown content from
raw.githubusercontent.com/anthropics/skills/using theWebFetchtool. - Boundary markers: The workflow relies on the
AskUserQuestiontool to clarify user intent and verify skill types before creation. - Capability inventory: Employs
Write,Bash,Read, andGlobtools to create files and interact with the local filesystem inSKILL.md. - Sanitization: Content is sourced exclusively from a trusted repository to guide local skill structure.
Audit Metadata