golang-popular-libraries

Pass

Audited by Gen Agent Trust Hub on Sep 14, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill utilizes WebFetch, WebSearch, and documentation tools to gather information from external sources such as pkg.go.dev and GitHub. This behavior creates a potential surface for indirect prompt injection where instructions hidden in third-party documentation could attempt to influence the agent's behavior during the vetting process. * Ingestion points: The SKILL.md file enables the use of WebFetch, WebSearch, and mcp__context7__query-docs to ingest external content. * Boundary markers: The skill does not provide specific instructions or delimiters to isolate external content from the agent's core instructions. * Capability inventory: The agent is authorized to execute Go-related commands (Bash(go:*)), Git operations, and file system modifications (Write, Edit). * Sanitization: There are no defined requirements for sanitizing or filtering data retrieved from external URLs before it is analyzed.
  • [EXTERNAL_DOWNLOADS]: The skill's reference files (libraries.md, tools.md) include numerous links to external Go libraries and development tools hosted on GitHub. These references point to widely recognized and established projects (e.g., maintained by Google, Uber, and the official Go team) and are central to the skill's primary function of dependency recommendation.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 14, 2026, 05:10 AM
Security Audit — agent-trust-hub — golang-popular-libraries