Browse reusable Agent Skills, each with a clear purpose and practical guidance.
Enable VK_LAYER_KHRONOS_validation on the erhe Quest APK so silent GPU hangs become loud aborts. Covers the bundle (default on), the runtime config knob, the on-device SPIR-V cache wipe via uninstall, and the device-error abort hook in editor.cpp. Use when a Quest build runs but renders nothing in VR, hangs at the Horizon launch panel, or otherwise misbehaves silently.
Invoke IMMEDIATELY via python script to stress-test decisions and reasoning. Do NOT analyze first - the script orchestrates the critique workflow.
Run ServiceRadar web-ng locally against the live Kubernetes `demo` CNPG database for dashboard, SRQL, services, and UI testing. Use when a task asks to test with live demo data, run local Phoenix against demo CNPG, validate dashboard authoring with Playwright, or avoid fragile kubectl port-forward database access.
Run ServiceRadar web-ng locally against the Kubernetes demo namespace FieldSurvey data, including CNPG NodePort access, NATS Object Store artifact access, authenticated browser checks, and Playwright screenshots for dashboard and FieldSurvey review iteration.
Run ServiceRadar Elixir database tests against the Kubernetes CNPG instance in the `srql-fixtures` namespace. Use when local localhost PostgreSQL is unavailable, when tests need current branch migrations, or when a user asks to use the srql-fixtures database for DB-backed validation. Covers scratch database creation, TLS-required NodePort connections, migrations, focused test commands, cleanup, and secret hygiene.
Run ServiceRadar `elixir/web-ng` locally against the Docker Compose CNPG database with copied mTLS certs and Docker secrets, then verify dashboard UI changes with Playwright. Use when iterating on web-ng, dashboard package, custom React dashboard, UAL WiFi map, SRQL topbar, or browser screenshot behavior without rebuilding the web-ng release image.
Manually test a running Vellum assistant end-to-end purely from the CLI — no desktop app or web UI. Hatch an instance, send messages, watch the reply, and tear it down. Use when verifying assistant behavior, reproducing a bug, or smoke-testing a change without the macOS/web clients.
Test TypeScript code snippets before persisting as skills
Review Vellum Assistant code changes for correctness, repo-specific quality rules, security risks, and missing validation. Use when reviewing diffs, preparing a PR, finishing implementation work, or when the user asks for a code review, quality pass, or pre-merge check in this repository.
Validate Vellum Assistant database and workspace migrations. Use when adding, editing, reviewing, or testing migrations, release-note migrations, persisted schemas, workspace file formats, or data backfills.