npm-downloads-to-leads
Pass
Audited by Gen Agent Trust Hub on May 15, 2026
Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
- [COMMAND_EXECUTION]: The skill uses Python shell blocks and a local script (
scripts/fetch.py) to process data and calculate growth metrics. This is the intended behavior for lead generation and scoring. - [EXTERNAL_DOWNLOADS]: The skill fetches package metadata and download counts from official, well-known services (api.npmjs.org and api.github.com). These network operations are necessary for the skill's primary function and do not involve downloading executable code from untrusted sources.
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data from the npm registry (package descriptions) and GitHub (user bios) to generate outreach messages. This is a common attack surface for indirect injection, though the risk is minimized by the skill's specific purpose of generating passive text for human review.
- Ingestion points: Maintainer bios and package descriptions are fetched via API in
scripts/fetch.pyand processed in Step 6 ofSKILL.md. - Boundary markers: None explicitly used to wrap external content during the generation of the lead brief.
- Capability inventory: The skill has the ability to execute Python code, write files to the local
docs/directory, and perform network requests to npm and GitHub. - Sanitization: A self-QA step (Step 7) is implemented to filter forbidden words and verify data consistency before output.
Audit Metadata