update-skills

Pass

Audited by Gen Agent Trust Hub on Jun 13, 2026

Risk Level: SAFE
Full Analysis
  • [COMMAND_EXECUTION]: The skill uses a suite of local shell utilities including git, fd, find, ripgrep (rg), and jq to perform automated file discovery and verify the content of instruction files.
  • [COMMAND_EXECUTION]: The workflow involves running just (a command runner) with the --summary flag to list available project tasks and potentially calling project-defined formatters like prettier to clean up updated files.
  • [DATA_EXFILTRATION]: The skill accesses local repository configuration files, such as package.json, justfile, and environment variable templates, to verify the technical accuracy of documentation. All data access is strictly limited to the project's local directory tree.
  • [EXTERNAL_DOWNLOADS]: The skill may invoke prettier via the NPM package manager (nlx/npx) only if it is already present as a project dependency, ensuring any network interaction is limited to a trusted registry for an existing tool.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 13, 2026, 09:57 AM
Security Audit — agent-trust-hub — update-skills