AGENT SKILL

frontend-development

Modify Langflow's React/Vite frontend, graph workspace, stores, icons, frontend tests, and UI/backend contract touchpoints.

License 许可
MIT
最近更新
2026-09-03
Stars
★ 351
Forks
27

01 前言 Front Matter

name
frontend-development
license
MIT
metadata
{ "disco-role": "operating" }
description
Modify Langflow's React/Vite frontend, graph workspace, stores, icons, frontend tests, and UI/backend contract touchpoints.
disable-model-invocation
true

02 资源文件 4

03 SKILL.md 正文

Frontend Development

Use this sub-skill when a task changes Langflow's browser UI: the visual flow workspace, React components, Zustand stores, API controllers/hooks, frontend tests, custom icons, Vite build configuration, or UI/backend data contracts.

Route Python API/service implementation to ../backend-runtime/SKILL.md. Route Python component class semantics, component indexes, and backend component compatibility to ../component-development/SKILL.md; this sub-skill covers the frontend consequences of those backend changes.

Start Here

  1. Confirm Node.js is at least 20.19.0 and npm is available. The frontend package is private and is built with React, TypeScript, Vite, Biome, Jest, Playwright, @xyflow/react, Axios, and Zustand.
  2. For local UI development, run the backend and frontend separately: start the backend on port 7860, then run the Vite frontend on port 3000. Set VITE_PROXY_TARGET=http://localhost:7860 when the frontend should proxy API calls to a non-default backend.
  3. Use references/frontend-architecture.md before touching graph nodes, stores, API controllers, or tests.
  4. Use references/ui-contracts-and-icons.md before changing backend/frontend contracts, icon names, lazy icon imports, or component palette display behavior.
  5. Use references/troubleshooting.md when npm install, Vite, Jest, Playwright, proxy, schema, icon, auth, or backend readiness failures appear.

Common Commands

Run frontend commands from the frontend package directory unless using a root make target.

  • Install dependencies as part of full repo setup: make init.
  • Start the dev frontend through the repo target: make frontend.
  • Start directly from the frontend package: npm start.
  • Build production assets: npm run build.
  • Run Jest unit tests: npm test or a targeted Jest invocation.
  • Run Playwright e2e tests when safe and dependencies are installed: npx playwright test tests/core --project=chromium.
  • Format and lint frontend code: make format_frontend, npm run check-format, npm run lint, and npm run type-check.

Bundled Helper

Use scripts/check_frontend_package.py for a safe static check of a frontend package.json before attempting npm commands:

python scripts/check_frontend_package.py src/frontend/package.json

The helper verifies that expected npm scripts such as start, build, test, and at least one quality gate are present without running Node, npm, Vite, Jest, or Playwright.

Task Checklist

  • Identify whether the change is UI-only, a UI/backend contract change, or a backend component display refresh; route backend semantics to the right sibling sub-skill before editing frontend assumptions.
  • Keep backend component class names stable. The frontend uses backend-provided component types/templates and can flag updates, but renaming a backend component class breaks saved flows.
  • Update the closest store, controller query, custom node helper, or UI component instead of adding cross-cutting state in unrelated files.
  • Add or update focused Jest tests for stores, controllers, hooks, and UI utilities; use Playwright for graph workspace or full browser regression behavior.
  • Validate with the narrowest relevant command first, then expand to lint/type/build or Playwright only when the environment supports it.

04 同仓库更多 Skills

VectorSpaceLab
VectorSpaceLab/AREX-Skill

passnet-triton-opt

Make a single PassNet Triton kernel fast AND numerically faithful: performance model (when a replacement can win at all), block/grid/warp tuning, autotune policy, launch overhead, and per-op numeric recipes to pass the dtype baseline tolerances. Use when a pass matches and is correct but speedup below expected, or when correctness fails by small numeric margins.

★ 351
VectorSpaceLab
VectorSpaceLab/AREX-Skill

passnet-solve

END-TO-END playbook for solving one PassNet sample: analyze the computation graphs, decide the optimization strategy (what to fuse, what to replace, what to leave alone), drive the iteration loop, and maximize the sample score. This is the ENTRY skill — invoke it first for any PassNet optimization task; it tells you when to use passnet-pattern-fusion, passnet-triton-opt and passnet-feedback.

★ 351
VectorSpaceLab
VectorSpaceLab/AREX-Skill

passnet-skill

PassNet GPU kernel optimization via compiler passes. Design and implement Triton-based optimization passes, create pass files under ./pass_dir/, self-evaluate with pass_evaluator, and iterate to maximize GPU speedup.

★ 351
VectorSpaceLab
VectorSpaceLab/AREX-Skill

passnet-pattern-fusion

Author PassNet pass files: write patterns that actually MATCH the FX graph, pick fusion regions (which ops to absorb into one Triton kernel), and structure multi-pass submissions with the mandatory shared-dispatch architecture. Use when creating or fixing pattern()/replacement_args()/replacement_func() files, when a pass "failed to match", or when deciding how to fuse multiple kernels/ops.

★ 351
VectorSpaceLab
VectorSpaceLab/AREX-Skill

passnet-orchestrate

Round/batch planning and decision support for PassNet graph optimization. Use for multi-sample triage, worker allocation, eval-budget policy, and post-eval keep/revert decisions. For solving one concrete sample end to end, use passnet-solve as the entry skill; consult this skill only when you need broader planning or a specific decision gate.

★ 351
VectorSpaceLab
VectorSpaceLab/AREX-Skill

passnet-feedback

Fast, deterministic feedback for PassNet work: pre-flight pattern/match verification WITHOUT burning a GPU evaluation, per-node bottleneck analysis of a sample's graphs, and eval-log parsing into per-variant status + estimated score + failure classification. Use BEFORE every GPU evaluation (check_pattern), at the START of a sample (analyze_graph), and AFTER every evaluation (parse_eval_log).

★ 351