What is agentic-ui AGENTS.md?

Contributor guidance for Agentic UI, a React and TypeScript component library for displaying AI-agent reasoning steps, tool calls, and task collaboration.

Use it to develop or test UI components, maintain translations, update documentation, and work with the project's React, Dumi, Father, Vitest, and Playwright setup.

It documents the project's conventions, supported environment, build tools, testing approach, and design goals before you change the library.

agentic-ui AGENTS.md — Instructions file by antdigital-ai

Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

agentmods

npx agentmods add instructions/antdigital-ai/agentic-ui/agents-md

Clone the repo

git clone --depth 1 https://github.com/antdigital-ai/agentic-ui

Made for: Codex, OpenCode.

Per session 9,544 This file is loaded in full into every session.

When invoked 9,544 The same file — it is already loaded in full.

Security scan C 2 findings. Scan, not verified.

Origin original No closer match found in the catalogue.

Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

Model	Per session	Once invoked
Fable 5	$0.09544	$0.09544
Opus 5	$0.04772	$0.04772
Sonnet 5	$0.01909	$0.01909
Haiku 4.5	$0.00954	$0.00954

Measured 3d ago against content hash fa5cba2b174b, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade C, and why

agentic-ui AGENTS.md scanned grade C with 2 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 3d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Recursive force deletehighDestructive command

rm -rf with a variable or a broad path is one typo away from removing the wrong tree.

rm -rf node_modules pnpm-lock.yaml

Makes network callslowCapability

Not a fault in itself. Listed so you know the mod talks to something, and to what.

- **首屏编译延迟（非 bug）**：dumi/umi 是 SPA，首次请求会触发按需编译，初始 HTML 仅返回 `Umi Loading...` 外壳，需等待数秒前端才渲染出内容；`curl` 拿到 200 不代表页面已可交互。

AGENTS.md · 969 lines

How it starts

The opening of the file, as written. The whole thing — 969 lines — stays where its author put it; the contents beside it link to each section on GitHub.

AGENTS.md

Agentic UI 项目开发指南 - 为 AI 编程助手提供项目上下文和开发规范

项目背景

这是 ant-design/agentic-ui 的源代码仓库，发布为 npm 包 @ant-design/agentic-ui，是一个面向智能体的 UI 组件库。

核心特性

使用 TypeScript 和 React 开发
兼容 React 16.9+ 版本（peerDependencies: >=16.9.0）
基于 Ant Design 和 @ant-design/cssinjs 构建
提供多步推理可视化、工具调用展示、任务执行协同等 Agentic UI 能力
支持国际化（i18n）
使用 Dumi 构建文档站点
使用 Father 构建产物
使用 Vitest + React Testing Library 进行单元测试
使用 Playwright 进行 E2E 测试

设计理念

过程透明化：可见思考与工具调用，让用户理解智能体的推理过程
主动协作：智能体主动发起交互，与用户协同完成任务
端到端任务协同：从"回答一句话"到"完成一件事"，让智能体成为用户的协作伙伴

快速开始

开发环境要求

Node.js: >= 18.17.0（推荐 22 LTS）
包管理器: 仓库锁定 pnpm 9.15.9（与 pnpm-lock.yaml lockfile v9 一致；勿用全局 pnpm 10+，其在 Node 18 上会报 requires at least Node.js v22.13）
操作系统: Windows 10+, macOS 10.15+, Linux
浏览器兼容性: 现代浏览器（Chrome 80+、Edge、Firefox、Safari）

安装依赖

# 克隆项目
git clone [email protected]:ant-design/agentic-ui.git
cd agentic-ui

# 使用 Corepack 启用 package.json 中锁定的 pnpm 版本（Node 16.9+ 自带 Corepack）
corepack enable
pnpm install

若未使用 Corepack，可显式激活同一版本：

corepack enable
corepack prepare [email protected] --activate
pnpm install

常用开发命令

pnpm start              # 启动文档站点（http://localhost:8000）
pnpm run build          # 构建项目
pnpm test               # 运行单元测试（默认跳过 e2e、大体积 chart/Workspace 目录、branches/coverage 等补洞套件，见 vitest.config.ts）
pnpm run test:full      # 全量 Vitest（`--mode full`；与 `test:coverage:full` 同一套 exclude，不含覆盖率）
pnpm run test:coverage  # 生成覆盖率（默认同上精简集）
pnpm run test:coverage:full  # 全量测试 + 覆盖率（`--mode full`；Codecov workflow / 发布前）
pnpm run test:e2e       # 运行 E2E 测试
pnpm run lint           # 代码检查（ESLint + Stylelint）
pnpm run prettier       # 代码格式化
pnpm tsc                # TypeScript 类型检查

Read the full file on GitHub · 969 lines

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

3d ago First seen · 969 lines · 9,544 tokens per session scan C fa5cba2b174b

Subscribe to this mod's changes

agentic-ui AGENTS.md is an instructions file published in the GitHub repository antdigital-ai/agentic-ui (224 stars, last pushed 21d ago), licensed MIT. It adds 9,544 tokens to every session, about $0.0477 per session on Opus 5. A static security scan graded it C with 2 findings (recursive force delete, makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.

vscode buildNext.instructions.md

Working notes and architecture documentation for the new esbuild-based build system in build/next. Use when making changes to the new build pipeline (transpile/bundle commands, NLS plugin, source-map handling, resource copying, or self-hosting watch tasks).

microsoft/vscode · 6,785 tokens

agentic-ui AGENTS.md