skills/simota/agent-skills/launch/Gen Agent Trust Hub

launch

Pass

Audited by Gen Agent Trust Hub on Oct 28, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [COMMAND_EXECUTION]: The skill uses the GitHub CLI (gh) extensively for fetching pull request data, managing releases, and checking rate limits. This is the primary mechanism for its reporting and orchestration features.
  • [EXTERNAL_DOWNLOADS]: The skill references and downloads well-known developer tools and libraries from trusted or established services. Specifically, it fetches chart.js from cdn.jsdelivr.net for report visualization and references standard automation tools like semantic-release, release-please, and git-cliff. These are documented neutrally as part of the intended release automation workflow.
  • [DYNAMIC_EXECUTION]: The skill includes local JavaScript and Shell scripts (generate-report.js, html-to-pdf.sh, puppeteer-pdf.js) to process data and export reports. These operate on local templates and data fetched via gh and do not execute untrusted remote code.
  • [INDIRECT_PROMPT_INJECTION]: The skill acts as a surface for processing pull request titles and descriptions. It follows best practices by acknowledging that AI-generated code or untrusted PR data requires explicit verification gates and does not grant the agent deployment permissions, maintaining a 'read-only' posture for reporting.
Audit Metadata
Risk Level
SAFE
Analyzed
Oct 28, 2026, 01:20 PM