researching-libraries

Pass

Audited by Gen Agent Trust Hub on Mar 6, 2026

Risk Level: SAFE
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill facilitates fetching metadata and documentation from established registries and services including npm (npms.io), PyPI, Go (pkg.go.dev), Rust (crates.io), and GitHub. These references are used solely for information gathering and library evaluation.
  • [COMMAND_EXECUTION]: Provides instructions for utilizing standard security auditing tools such as npm audit, pip-audit, govulncheck, and snyk. These commands are recommended for the legitimate purpose of identifying vulnerabilities in potential dependencies.
  • [PROMPT_INJECTION]: The instructional content focuses on enforcing a research-first development policy. There are no patterns suggesting attempts to bypass safety filters, extract system prompts, or override agent behavioral constraints.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 6, 2026, 02:12 PM