lov-bp
Pass
Audited by Gen Agent Trust Hub on Sep 3, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process external project materials such as PRDs, customer notes, and websites to generate business plan content. Maliciously crafted content within these documents could potentially influence the agent's behavior or the narrative of the generated BP.
- Ingestion points: materials referenced in
lov-bp-outline/SKILL.md(repositories, product docs, PRDs, websites, analytics exports). - Boundary markers: The kit utilizes an 'Evidence Ledger' (
evidence-ledger.md) to categorize claims as facts, inferences, or assumptions, providing a structure to verify ingested data. - Capability inventory: Writing workspace files, generating audit reports, and delegating slide generation to sub-skills.
- Sanitization: The provided Python scripts do not implement explicit sanitization or filtering for external content interpolated into templates.
- [COMMAND_EXECUTION]: The kit includes and utilizes local Python scripts (
scripts/init_bp.py,scripts/audit_bp.py) to initialize the project workspace and perform structural linting on the narrative. These scripts perform file system operations, such as creating directories and writing files, based on user-supplied parameters and environment variables.
Audit Metadata