golang-pkg-go-dev

Pass

Audited by Gen Agent Trust Hub on Oct 15, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill requires the installation of a third-party command-line tool, godig, from a repository not associated with the provided trusted organization list.
  • Evidence: Instructions recommend running go install github.com/samber/godig/cmd/godig@latest to enable the skill's functionality.
  • [INDIRECT_PROMPT_INJECTION]: The skill facilitates the ingestion of content from the public Go package ecosystem (pkg.go.dev), creating a surface for indirect prompt injection.
  • Ingestion points: The skill fetches full package documentation, module READMEs, and runnable examples using commands like package doc, module readme, and package examples from SKILL.md.
  • Boundary markers: There are no specific boundary markers or instructions to treat the retrieved documentation as untrusted data that should not be executed or obeyed.
  • Capability inventory: The agent, upon receiving instructions from malicious documentation, could potentially use other skills or tools in its environment to perform unauthorized actions.
  • Sanitization: No sanitization or filtering of the fetched content is implemented; the data is passed directly to the agent in Markdown or raw format.
Audit Metadata
Risk Level
SAFE
Analyzed
Oct 15, 2026, 09:23 AM