create-website
Pass
Audited by Gen Agent Trust Hub on Aug 28, 2026
Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill instructs the agent to suggest or perform shell commands using
npx skills addto install additional modular dependencies. - [EXTERNAL_DOWNLOADS]: Operating Rule 4 recommends downloading external skill components from the author's registry (
wondelai/skills). - [INDIRECT_PROMPT_INJECTION]: The skill reads existing files from the
docs/directory to maintain and resume journey state, which creates a potential vector for instructions embedded in those files to influence agent behavior. - Ingestion points: Files in the
docs/directory (e.g.,docs/CREATE-WEBSITE-PLAN.md) are read during the initial 'Resume first' phase. - Boundary markers: The skill does not implement delimiters or safety instructions to isolate content read from external artifacts.
- Capability inventory: The skill has the ability to write to the file system and suggest command execution.
- Sanitization: Ingested content is summarized and processed without explicit validation or escaping of the raw file data.
Audit Metadata