Agent skills

Use when a request has material product, behavior, or architecture choices that must be resolved before implementation

Use when planning or storyboarding an AI video or video ad before generating it — building a 10-frame / 15-second "contact sheet"…

Structured medical case presentation for clinical rounds, conferences, and documentation.

Quarterly / monthly financial report — masthead with KPIs, revenue and burn charts, P&L summary table, top-line highlights, and an outlook…

Short-form video generation skill — 3-10 second clips for product reveals, motion teasers, ambient loops.

Make a Bun/TS command-line tool's output look designed — the render layer, in the house style used by `qb` and `see`.

Captures polished screenshots or video of a CLI, REPL, or text-based tool in a terminal.

Pages: Terms of Service Guides Terms of Service page content, structure, compliance, and AI-specific clauses.

A sun-baked, clay-toned editorial interface built on warm cream surfaces, ink-brown headlines set in a display serif, and a single…

Minimize code and maintenance surface while preserving correctness, clarity, safety, and requested behavior.

Test FastMCP servers through direct unit tests, in-memory MCP clients, elicitation handlers, and focused deployed-transport integration…

Test a running HTTP API end-to-end with hurl — write .hurl files with request + assert blocks, handle bearer-token auth and variables, run…

Run tests and auto-fix failures. Standalone test runner that diagnoses and repairs failing tests without manual intervention.

Detect and fix testing anti-patterns for better test quality

Invoke whenever writing, changing, reviewing, or sweeping tests.

Use when generating test cases after a solution is verified solvable, designing adversarial tests to break wrong approaches, creating…

Orchestrate parallel test generation for all changed components, jobs, models, policies, and services in the current branch.

Analyze test coverage and suggest tests for uncovered code

Use when test coverage falls below target or a new module has no tests — identifies untested code paths and writes targeted unit or…

Use when testing the Hubble Electron desktop app, especially when inspecting, clicking, screenshotting, or verifying a real note edit in…

Walk one test-driven development cycle (RED then GREEN, optionally Refactor while GREEN).