lov-bp-outline
Pass
Audited by Gen Agent Trust Hub on Oct 8, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill instructs the agent to execute a bundled Python script
scripts/init_bp.pyto set up the project workspace, create directories, and populate templates using provided project parameters. - [INDIRECT_PROMPT_INJECTION]: The skill processes external project materials, including repositories, product documentation, websites, and customer notes, to extract evidence for the business plan. This creates a surface for indirect prompt injection where malicious instructions embedded in those documents could influence the agent's behavior.
- Ingestion points: Supply scope search (Step 1) which involves reading repository files, PRDs, and external website content.
- Boundary markers: No specific delimiting characters or instructions are provided to the model to ignore potential instructions embedded within the ingested materials.
- Capability inventory: The skill possesses the capability to execute a local initialization script and perform file system writes within the workspace directory.
- Sanitization: The skill does not perform explicit sanitization or filtering of external document content before it is interpolated into the reasoning process.
Audit Metadata