lov-bp-outline
Pass
Audited by Gen Agent Trust Hub on Oct 23, 2026
Risk Level: SAFE
Full Analysis
- [COMMAND_EXECUTION]: The skill executes a local script
scripts/init_bp.pyto set up the project workspace. This script uses standard Python libraries to create directories and write template files based on user input. - [DATA_EXPOSURE]: The initialization script accesses the user's home directory to read a configuration file (
~/.skill-publisher/skills/profile.json). This is used to resolve user preferences and brand settings, which is consistent with the skill's stated purpose of managing portable project configurations. - [INDIRECT_PROMPT_INJECTION]: The skill processes external project materials such as PRDs, documentation, and customer notes which could contain untrusted content. The skill mitigates this by implementing a mandatory 'evidence gate' workflow where every claim must be categorized as a fact, inference, or assumption and verified against specific source IDs.
- [SAFE]: The skill includes defensive logic in
scripts/init_bp.pyto prevent writing to restricted system paths (like the root or home directories). No network operations, obfuscation, or persistence mechanisms were detected.
Audit Metadata