scaffold-course
Pass
Audited by Gen Agent Trust Hub on Jun 18, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill is vulnerable to indirect prompt injection via the 'slug' and 'title' arguments.
- Ingestion points: The '$ARGUMENTS' section in 'SKILL.md' identifies 'slug' and 'title' as user-controlled inputs.
- Boundary markers: No explicit delimiters or 'ignore embedded instructions' warnings are provided to the agent to treat these inputs as untrusted data.
- Capability inventory: The skill possesses file-writing capabilities, specifically creating TypeScript scripts in '{{SCRIPTS_DIR}}' and markdown files in 'content-library/'. It also instructs the agent to create scripts that interact with the database via Drizzle.
- Sanitization: There is no instruction to sanitize the 'slug' for path traversal characters (e.g., '../') or to escape the 'title' and 'slug' when interpolating them into the generated TypeScript code or markdown headers. This could lead to arbitrary file creation outside the intended directory or code injection within the generated ingestion script.
Audit Metadata