Browse reusable Agent Skills, each with a clear purpose and practical guidance.
Tune literal `register_cuda_ci(est_time=...)` values in `tests/e2e` from complete GitHub Actions logs produced by `ci-fetch-log`. Use when asked to calibrate E2E timing estimates, produce an evidence-backed runtime CSV, or apply estimate-only changes. Aggregate passed samples by test file, apply a 1.25 safety factor with mandatory upward bucket rounding, and never commit or push changes.
Fetch complete GitHub Actions job logs with `gh`, write an auditable completeness manifest, and optionally diagnose failed jobs from saved evidence. Use when asked to collect full CI logs, save a workflow run for later analysis, or explain failed jobs. A PR number resolves to the newest workflow run for the PR's current head SHA. PR numbers, run URLs, and job URLs default to the whole run; `--only` requires an explicit single-job request.
Release-gate scan for deliberately planted malicious code (crypto miners, data/secret exfiltration, backdoors, RCE/obfuscation, install hooks) across the whole working tree of World of ClaudeCraft. Use before tagging or shipping a release, or whenever you want to confirm the tree is free of malicious code. Runs the deterministic scanner, fans the read-only release-malware-audit agent across categories, and returns a single PASS / BLOCK verdict with confirmed findings and dismissed false positives. Distinct from privacy-security-review, which catches accidental security mistakes.
Prepare, validate, build, and use Nemotron Customizer airgap image bundles for offline clusters. Use when planning airgapped deployments, editing deploy/nemotron-customizer/airgap/airgap.yaml, selecting workflow targets, grouping step execution images, baking repo overlays or wheel additions, resuming airgap runner builds, or submitting `nemotron steps run` jobs inside an airgapped environment.
Build and test the Electron app with proper worktree isolation
Use when the user wants to author, design, or test an Azure Policy that queries Azure Resource Graph (ARG) at request-time — i.e. a policy whose deny/audit decision depends on data from elsewhere in the subscription (sibling/parent resource state, RG-wide invariants, multi-hop relationships, etc.). Formally called Azure Policy External Evaluation; sometimes referred to colloquially as "Invoke". Drives an iterative KQL co-design loop against the user's real subscription via `az graph query`, then emits a policy definition, assignment, `.http` test flow, and an `EXPLANATION.md` companion. Read-only; never provisions anything.
Show the current migration status for all configured databases
Rollback the last applied database migration (rockhopper down)