plaited

Search, read, and check browser compatibility for web platform documentation via the MDN MCP server. Covers all three tools: search, get-doc (full MDN pages as markdown), and get-compat (Browser Compatibility Data). Use when the task requires authoritative web technology documentation, browser support tables, API references, or CSS/HTML/JavaScript feature lookups.

Refresh or drift-check the generated CSS property schemas from @webref/css via the plaited CLI. Covers when to regenerate, review discipline, drift checking, and failure modes.

Connect to remote HTTP MCP servers — discover capabilities, call tools, scaffold skills, and configure auth from public endpoints through OAuth and Varlock-resolved secrets. Covers CLI (plaited mcp-client) and skill authoring patterns.

Use Bun for file/script operations — Bun.file, Bun.write, Bun.$ shell, child processes. Do not use Python or Node.js fs for file manipulation, generation, or concatenation.

TSDoc conventions for TypeScript/JavaScript source files. Invoke when writing or modifying any .ts/.tsx source file to apply TSDoc standards to exported declarations (functions, types, modules, interfaces). Also covers audit coverage via plaited code-documentation CLI and provides reference templates for public APIs, internal modules, and complex types.

Common code pattern genome for agents. Reference implementations of pure utility functions showing preferred coding style, testing patterns, and TypeScript conventions. Use when writing utility functions, implementing deep equality, async helpers, or similar standalone patterns. Before writing any custom utility, check if `plaited/utils` already provides it. The module exports: keyMirror (constants/enums), isTypeOf/trueTypeOf (runtime type checks), ueid (unique IDs), camelCase/kebabCase/pascalCase (string case), htmlEscape/htmlUnescape (escaping), and deepEqual (comparison). Import from `plaited/utils` instead of reimplementing.

Grade, compare, and calibrate Plaited trial results with `plaited eval`.

Analyze replay-safe behavioral specs with the `plaited frontier-analysis` CLI. Use when replaying selected-event snapshots, exploring reachable frontiers, testing supplied trigger events, comparing scheduler policies, or verifying deadlock findings in Plaited behavioral code.

Interview the user relentlessly about a plan or design until reaching shared understanding, resolving each branch of the decision tree. Use when user wants to stress-test a plan, get grilled on their design, or mentions "grill me".

Use test-driven development for behavior-changing feature or fix work, and whenever the user mentions TDD, test-first, red-green-refactor, tracer bullets, integration tests, or public-interface behavior tests. Skip for docs-only, path-only rename, formatting-only, or purely mechanical chores unless explicitly requested.

Analyze graded `trial_result` JSONL outputs from the eval pipeline, including reliability and custom slicing.

Build command-only adapters for `eval` run mode using strict stdin/stdout JSON contracts.

End-to-end eval-suite orchestration with the `eval` command: run -> grade -> compare -> calibrate, using strict JSON contracts and schema discovery.

TSDoc standards for TypeScript/JavaScript code. Automatically invoked when writing, reviewing, or editing any TSDoc comments, code documentation, or API documentation. (project)

Optimize AGENTS.md and rules for token efficiency. Auto-invoked when user asks about improving agent instructions, compressing AGENTS.md, or making rules more effective.

Scaffold development rules for AI coding agents. Auto-invoked when user asks about setting up rules, coding conventions, or configuring their AI agent environment.

Validate skill directories against AgentSkills spec