Use whenever C++ Unreal Engine code needs tests that use the engine, editor or UMG UI: the Automation Test Framework's simple and complex tests, Automation Spec (Describe, It, latent tests), latent commands, the Automation Driver for UI input, Session Frontend and command-line runs in CI.

Use before a multiplayer game goes to platform cert, where broken multiplayer is a common rejection. Builds a coverage matrix for connect, authority handoff, disconnect, reconnect and host migration under latency, loss and reordering for Unity Netcode, Unreal and Mirror, with a go/no-go gate.

Use whenever GDScript code in a Godot project needs unit tests with GUT: install, tests extending GutTest, assertions including signals, lifecycle hooks, parameterized tests, doubles, stubs and spies, async tests, the command-line runner with JUnit XML, and CI. GUT 9 for Godot 4, GUT 7 for Godot 3.

Use when embedded tests must run on a virtual MCU or SoC instead of the host or real hardware, for correct endianness, alignment and interrupts: qemu-system-arm, aarch64 and riscv32 machine and CPU flags, loading ELF binaries, semihosting exit codes, GDB and QMP, and CI.

Use when deciding what code coverage level embedded C/C++ needs (statement, branch, MC/DC) and wiring gcov or llvm-cov into the build: host vs QEMU instrumentation, MISRA-C and DO-178C expectations by safety level, and report formats.

Use whenever a schema is managed with Flyway: versioned, repeatable and undo migration files, migrate, info, validate, baseline and repair, the schema history table, migration ordering questions, and CI gates on schema changes, via CLI, Maven, Gradle or Docker.

Use whenever two services (a frontend and an API, or two microservices) keep breaking each other's API: set up Pact consumer-driven contract tests, publish pacts to a broker, verify the provider and gate deploys with can-i-deploy.

Use when a multi-language codebase needs one complexity gate, or a function has grown unreviewable and CI needs a number. Runs Lizard (30+ languages) on production code to enforce per-function cyclomatic complexity, length and parameter-count thresholds, excluding tests.

Use when a chaos experiment is already designed and a fault is about to be injected into a running system. Sets the go/no-go gates (non-production, healthy baseline, live monitoring, tested rollback), blast-radius bound, abort criteria, inject and abort commands and the recovery check.

Use when a Cucumber, Behave or Reqnroll project has grown past ~50 step definitions, before writing a step that may already exist, or when new .feature files need a duplicate check. Inventories steps, merges duplicates, publishes a step-library README and a scenario coverage map.

Use whenever an API has an OpenAPI or GraphQL schema and needs tests generated from it: fuzzing every endpoint for 5xx errors and schema violations. Runs Schemathesis from the CLI or pytest, picks the conformance checks and gates CI on failures as the API evolves.

Use when an existing API test suite should be rerun under network faults to prove retries, timeouts, circuit breakers and fallbacks work. Injects latency, timeouts, dropped connections, bandwidth caps and packet loss with Toxiproxy and reports which assertions break under which fault.

Use when building or reviewing keyboard-only support or a modal, drawer or popover's focus handling: WCAG 2.2 keyboard access, no keyboard traps, shortcuts, focus order, visible and unobscured focus, with test scripts and the focus-trap pattern (focus on open, Escape, restore).

Fetches and analyzes GitLab merge request (MR) comments, metadata, and review feedback using authenticated API calls. Capabilities include fetching comment threads, retrieving reviewer feedback, filtering unresolved discussions, and generating MR activity reports. Use when the user asks about GitLab MR comments, code review discussions, review feedback, approval status, troubleshooting review delays, or needs to fetch merge request metadata via the GitLab API.

Use the connected scholar-sidekick-mcp MCP server when the user mentions a scholarly identifier (DOI including shortDOI aliases, PMID, PMCID, ISBN, arXiv, ISSN, NASA ADS bibcode, WHO IRIS URL) and wants structured metadata, a formatted citation, a bibliography export file, a retraction check, an open-access check, verification that a claimed citation is real (not fabricated), or a whole-bibliography audit (batch fabrication + retraction check). Requires the MCP server connected — it works anonymously, no API key required; for a zero-install path use the scholar-sidekick-api skill instead.

Resolve scholarly identifiers (DOI including shortDOI aliases, PMID, PMCID, ISBN, arXiv, ISSN, ADS bibcode, WHO IRIS URL) into formatted citations (10,000+ CSL styles) and bibliography exports (BibTeX, RIS, EndNote, CSV…), and check retraction, open-access, and citation-fabrication status — from the terminal via the `scholar` CLI. Wraps the public REST API; needs Node ≥20 but no API key for the free tier. Use when the user wants to look up, cite, export, or verify a reference by DOI, PMID, PMCID, ISBN, arXiv ID, ISSN, ADS bibcode, or WHO IRIS URL from the terminal — including retraction, open-access, and fabricated-citation checks, or a whole-bibliography audit of a .bib/.ris/CSL-JSON file.

Resolve scholarly identifiers (DOI including shortDOI aliases, PMID, PMCID, ISBN, arXiv, ISSN, ADS bibcode, WHO IRIS URL) into formatted citations (10,000+ CSL styles) and bibliography exports (BibTeX, RIS, EndNote, CSV…), and check retraction, open-access, and citation-fabrication status — for one citation or a whole bibliography at once. Calls a documented REST API over plain HTTP — no install, no API key needed for the free tier. Use when the user has a DOI, PMID, PMCID, ISBN, arXiv ID, ISSN, ADS bibcode, or WHO IRIS URL and asks to format a citation, build or export a bibliography, check whether a paper is retracted, open-access, or a fabricated citation, or audit every reference in a .bib/.ris/CSL-JSON bibliography.

Use the Agent Ready (agent-ready.dev) REST API to scan any public URL for AI agent-readability against the Vercel Agent Readability Spec, the llmstxt.org standard, and agent-protocol manifests (MCP server cards, A2A, agents.json, agent-permissions.json, UCP, x402, NLWeb). Activates for "scan this site for AI agent-readability", "run an Agent Ready scan on {URL}", "check the Agent Ready score for {URL}", "what's the agent-readability rating for {URL}", or any time the user wants a programmatic readability scan via HTTP. Picks this skill when the user does NOT have the Agent Ready MCP server installed — for MCP, use the `agent-ready-mcp` skill instead.

Build Kopai observability dashboards from OpenTelemetry metrics, logs, and traces by composing tiles — time-series charts, stat/KPI numbers, histograms, log timelines, and trace-detail views — into a dashboard via the Kopai SDK. Use this skill whenever the user wants to visualize, chart, monitor, or 'keep an eye on' telemetry — e.g. 'make a dashboard', 'monitoring view', 'KPI panel', 'chart request latency and throughput', 'an incident board with error logs and a trace viewer' — even if they don't say the word 'dashboard' — and when another workflow needs to present telemetry visually (e.g. after a root-cause analysis). This creates Kopai dashboards specifically. Do NOT use it to investigate or root-cause an issue (use root-cause-analysis), to add instrumentation (use otel-instrumentation), to build Grafana/Prometheus dashboards, or to code React/UI chart components. When the Kopai connector is attached, use create-live-page instead unless the user asks for a Kopai dashboard.

Investigate and root-cause production issues from live OpenTelemetry telemetry (traces, logs, metrics) with the Kopai SDK in TypeScript code mode. Use this skill to debug errors and error-rate spikes, investigate latency/slowness and timeouts, trace requests across services, find failing or cascading services, and correlate logs to a trace — including vague symptom reports like 'why is my API slow', 'getting 500 errors', 'service is down', 'requests are timing out', or 'something is failing in prod', even when the user never says 'traces' or 'observability'. This analyzes existing telemetry to find a cause. Do NOT use it to add instrumentation (use otel-instrumentation), to build dashboards or visualizations (use create-dashboard), or to fix non-telemetry problems like failing CI, lint errors, unit tests, or SQL query tuning.