plaited
- 11 skills
- 0 followers
- 21 hours ago last updated
- ▌ Markdown · plaitedParse, link-check, and extract YAML frontmatter from markdown documents via the `plaited markdown` CLI. Three modes — extract-links, validate-links (with rootRelative for bundle-relative `/` links), and frontmatter. Use when resolving local references, validating a documentation tree's cross-links, or reading metadata before editing docs.
- ▌ Git Context · plaitedGather structured local Git context before editing, reviewing, or committing via the `plaited git-context` CLI. Four modes — status, history, worktrees, context — returning branch, HEAD, upstream, staged/unstaged files, merge-base, commits-since-base, and per-path history as JSON. Replaces chaining 8+ raw git commands.
- ▌ Typescript Lsp · plaitedLSP-style queries over TypeScript 7 native API — documentSymbol, hover, completion, definition. Two modes (execute and discover) for semantic code analysis without a language-server process. Use when you need type info, symbol lists, definitions, or autocomplete for TypeScript/JavaScript code.
- ▌ Plaited Framework · plaited bundlePlaited behavioral-programming runtime and UI layer — b-threads, triggers, handlers, the controller/custom-element protocol, SSR/renderer, frontier analysis, behavioral eval capture, autoresearch hill-climbs, and OKF knowledge bundles. Use when creating, reading, updating, or deleting code where plaited is a declared dependency or where the work is in the Plaited repo itself.
- ▌
- ▌
- ▌
- ▌ Trial Adapters · plaitedBuild command-only adapters for `eval` run mode using strict stdin/stdout JSON contracts.
- ▌ Tdd · plaited bundleUse 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.
- ▌ Trial Runner · plaitedEnd-to-end eval-suite orchestration with the `eval` command: run -> grade -> compare -> calibrate, using strict JSON contracts and schema discovery.
- ▌ Compare Trials · plaited bundleAnalyze graded `trial_result` JSONL outputs from the eval pipeline, including reliability and custom slicing.