promote-idea
Pass
Audited by Gen Agent Trust Hub on Sep 14, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes data from an external file (
IDEAS.md) and interpolates it into new project documentation without explicit sanitization or boundary markers, creating a surface for indirect prompt injection. \n - Ingestion points: The agent ingests data from
IDEAS.mdusing theget_ideastool orReadcapability to extract idea concepts and metadata. \n - Boundary markers: The instructions do not specify the use of delimiters or 'ignore' instructions to isolate the imported concept text from the agent's executable context. \n
- Capability inventory: Through the
bitwize-music-mcptool, the skill has capabilities to create new directories (create_album_structure), modify existing tracking files (IDEAS.md), and write content into new project files (README.md). \n - Sanitization: The skill identifies and normalizes slugs for file system compatibility but does not perform sanitization on the thematic concept text being injected into the project files.
Audit Metadata