regression-testing
Pass
Audited by Gen Agent Trust Hub on Sep 18, 2026
Risk Level: SAFE
Full Analysis
- [COMMAND_EXECUTION]: The skill uses the GitHub CLI (
gh) to trigger workflows, monitor runs, and download artifacts. This is the primary intended functionality for CI/CD orchestration. - [EXTERNAL_DOWNLOADS]: It downloads test artifacts (logs, screenshots, traces) from the user's own GitHub Actions runs via
gh run download. - [INDIRECT_PROMPT_INJECTION]: The skill processes CI logs and test results as external data. This data is ingested for classification purposes (identifying regressions, flakiness, or environment issues). The skill uses sharding and subagents for this processing, which limits context exposure. The risk is assessed as low as the logs are treated as data, not instructions.
- [CREDENTIALS_UNSAFE]: The skill manages authentication tokens and API keys for external services (GitHub, Supabase, Vercel, Cloudflare). It follows safe practices by instructing the agent to set these in GitHub Secrets (
gh secret set) or local environment variables provided by the user, rather than hardcoding any credentials.
Audit Metadata