经 AI Skill Hub 精选评估,代码智能助手 获评「强烈推荐」。这款Agent工作流在功能完整性、社区活跃度和易用性方面表现出色,AI 评分 8.0 分,适合有一定技术背景的用户使用。
自动化AI工作流,读取项目、运行命令、写入代码
代码智能助手 是一套完整的 AI Agent 自动化工作流方案。通过可视化的节点编排,将复杂的多步骤任务拆解为清晰的自动化流程,实现全程无人值守的智能处理。支持与数百种外部服务和 API 无缝集成,适合构建数据处理管线、业务自动化和 AI 辅助决策系统。
自动化AI工作流,读取项目、运行命令、写入代码
代码智能助手 是一套完整的 AI Agent 自动化工作流方案。通过可视化的节点编排,将复杂的多步骤任务拆解为清晰的自动化流程,实现全程无人值守的智能处理。支持与数百种外部服务和 API 无缝集成,适合构建数据处理管线、业务自动化和 AI 辅助决策系统。
# 方式一:npm 全局安装 npm install -g codeep # 方式二:npx 直接运行(无需安装) npx codeep --help # 方式三:项目依赖安装 npm install codeep # 方式四:从源码运行 git clone https://github.com/VladoIvankovic/Codeep cd Codeep npm install npm start
# 命令行使用
codeep --help
# 基本用法
codeep [options] <input>
# Node.js 代码中使用
const codeep = require('codeep');
const result = await codeep.run(options);
console.log(result);
# codeep 配置说明 # 查看配置选项 codeep --config-example > config.yml # 常见配置项 # output_dir: ./output # log_level: info # workers: 4 # 环境变量(覆盖配置文件) export CODEEP_CONFIG="/path/to/config.yml"
<p align="center"> <img src="Codeep.svg" alt="Codeep Logo" width="200"> </p>
<p align="center"> <strong>Deep into Code.</strong> </p>
<p align="center"> <img src="Codeep.gif" alt="Codeep demo" width="700"> </p>
<p align="center"> Autonomous AI coding agent that reads your project, runs commands, and writes & verifies code — deeper than autocomplete. Any model (Claude, Gemini, DeepSeek, GLM, OpenAI, or your own local/custom), in the terminal and your editor. </p>
<p align="center"> <a href="https://www.npmjs.com/package/codeep"><img src="https://img.shields.io/npm/v/codeep.svg" alt="npm version"></a> <a href="https://www.npmjs.com/package/codeep"><img src="https://img.shields.io/npm/dm/codeep.svg" alt="npm downloads"></a> <a href="https://github.com/VladoIvankovic/Codeep/blob/main/LICENSE"><img src="https://img.shields.io/npm/l/codeep.svg" alt="license"></a> <a href="https://github.com/VladoIvankovic/Codeep"><img src="https://img.shields.io/github/stars/VladoIvankovic/Codeep?style=social" alt="GitHub stars"></a> </p>
/settings/tasks add and /tasks doneCodeep advertises the following ACP capabilities so Zed (and other ACP clients) can use them:
promptCapabilities.image — paste/drag images into chat for vision analysis.promptCapabilities.embeddedContext — drag a file or pin a code selection into the chat and the actual file content is injected into the prompt (resource & resource_link blocks, 200 KB cap per file).sessionCapabilities.list — exposes saved CLI sessions to ACP clients.sessionCapabilities.resume — instant reconnect on panel reload without replaying history.loadSession — restore a previous session by id.Codeep also reads clientCapabilities.terminal from the client and routes shell commands through the editor's terminal when supported (so you see them in Zed's terminal panel), falling back to local execution otherwise.
Run the test suite, build, then npm run deploy:staging. If the build fails, surface the error and stop — don't deploy a broken build. ```
The agent gets the bundle catalog injected into its system prompt on every iteration and can invoke any bundle via the invoke_skill tool:
/skills bundles — list installed bundles/skills create-bundle <name> — scaffold a new project skill/skills show <name> — print the SKILL.md/skills browse [query] — search the public marketplace/skills install <owner>/<slug> — pull from marketplace/skills publish <name> [--public] — share to codeep.dev/skills unpublish <owner>/<slug> — remove your published skillWeb marketplace: publish skills to codeep.dev/skills as public (browseable by anyone) or private (only you). Manage your published skills at /dashboard/skills. From the VS Code extension, use the Codeep: Browse Skill Bundles… / Codeep: Create Skill Bundle… / Codeep: Open Skills Folder commands.
npm install -g codeep
settings.json:"agent_servers": {
"Codeep": {
"type": "custom",
"command": "codeep",
"args": ["acp"]
}
}
export DEEPSEEK_API_KEY=your_key # or whichever provider
curl -fsSL https://github.com/VladoIvankovic/Codeep/releases/latest/download/codeep-macos-arm64 -o codeep chmod +x codeep sudo mv codeep /usr/local/bin/ ```
```bash
You write safe, reversible DB migrations… ```
tools is an allowlist enforced at dispatch (a researcher literally can't write files). Sub-agents inherit your profile + project rules, and their changes are covered by /undo.
Guaranteed review: enable Agent Auto-Review in /settings (agentAutoReview) and after any run that changes files, Codeep automatically delegates to the reviewer and appends its findings — a review stage that always happens. Off by default.
codeep-min-version: 2.0.0 codeep-requires-mcp: [postgres] allowed-tools: [read_file, write_file, execute_command]
When enabled in Settings, the agent can automatically verify its changes by running build, tests, or type checking after completing a task. Disabled by default — the agent works freely and you decide when to verify.
Configure in Settings → Agent Auto-Verify: - Off (default) — no automatic verification - Build only — checks compilation after changes - Typecheck only — runs TypeScript/PHP type checking - Test only — runs test suite after changes - Build + Typecheck + Test — full verification
If errors are found, the agent tries to fix them automatically (up to 3 attempts, configurable).
Supported project types:
| Language | Build | Test | Type Check |
|---|---|---|---|
| **Node.js/TypeScript** | npm/yarn/pnpm/bun run build | npm test, vitest, jest | tsc --noEmit |
| **Python** | - | pytest | - |
| **Go** | go build | go test | - |
| **Rust** | cargo build | cargo test | - |
| **PHP/Laravel** | composer run build | phpunit, artisan test | php -l (syntax) |
account sync / account push also carry your personalities and custom slash commands between machines — the Markdown files in ~/.codeep/personalities/ and ~/.codeep/commands/:
codeep account push # Upload local personalities + commands to codeep.dev
codeep account sync # Download them onto a new machine
Merging is additive — sync only writes files that don't already exist locally, so it never clobbers a personality or command you've edited on this machine. View what's synced (and prune stale entries) under Personalities and Custom commands on the dashboard.
Hooks and MCP server configs are deliberately not synced: hooks run arbitrary shell, and MCP configs often embed tokens, so both stay local to each machine.
curl -fsSL https://raw.githubusercontent.com/VladoIvankovic/Codeep/main/install.sh | bash
Custom installation directory:
curl -fsSL https://raw.githubusercontent.com/VladoIvankovic/Codeep/main/install.sh | INSTALL_DIR=~/.local/bin bash
Specific version:
curl -fsSL https://raw.githubusercontent.com/VladoIvankovic/Codeep/main/install.sh | VERSION=1.0.0 bash
brew tap VladoIvankovic/codeep
brew install codeep
Update:
brew upgrade codeep
npm install -g codeep
Update:
npm update -g codeep
Download the latest binary for your platform from GitHub Releases:
| Platform | Binary |
|---|---|
| macOS Apple Silicon (M1/M2/M3/M4) | codeep-macos-arm64 |
| macOS Intel | codeep-macos-x64 |
| Linux x86_64 | codeep-linux-x64 |
```bash
| Command | Description |
|---|---|
/provider | Switch AI provider |
/model | Switch AI model |
/model <name> | Load a saved profile shortcut (e.g. /model fast) |
/protocol | Switch API protocol (OpenAI/Anthropic) |
/lang | Set response language (12 languages supported) |
/settings | Adjust temperature, max tokens, timeout, rate limits |
/profile save <name> | Save current provider+model as a named profile |
/profile load <name> | Load a saved profile |
/profile <name> | Shorthand for /profile load <name> |
/profile list | List all saved profiles |
/profile delete <name> | Delete a saved profile |
/cost /stats | Show session token usage and estimated API cost per model |
Model favorites — save provider+model combos and switch instantly:
> /provider # switch to z.ai
> /model glm-5.1
> /profile save fast
> /provider # switch to openai
> /model gpt-4.1
> /profile save work
> /model fast # instantly switch to z.ai / glm-5.1
> /model work # instantly switch to openai / gpt-4.1
| Type | Location |
|---|---|
| Global config | ~/.config/codeep/config.json |
| Project config | .codeep/config.json |
| Global sessions | ~/.codeep/sessions/ |
| Project sessions | .codeep/sessions/ |
| Global logs | ~/.codeep/logs/ |
| Project logs | .codeep/logs/ |
| Variable | Description |
|---|---|
ZAI_API_KEY | Z.AI (international) API key |
ZAI_CN_API_KEY | Z.AI China API key |
OPENAI_API_KEY | OpenAI API key |
ANTHROPIC_API_KEY | Anthropic Claude API key |
DEEPSEEK_API_KEY | DeepSeek API key |
GOOGLE_API_KEY | Google AI (Gemini) API key |
MINIMAX_API_KEY | MiniMax (international) API key |
MINIMAX_CN_API_KEY | MiniMax China API key |
| Setting | Default | Description |
|---|---|---|
| Temperature | 0.7 | Response creativity (0.0 - 2.0) |
| Max Tokens | 8192 | Maximum response length |
| API Timeout | 60000ms | Request timeout |
| API Rate Limit | 30/min | Max API calls per minute |
| Command Rate Limit | 100/min | Max commands per minute |
| Agent Mode | ON | ON = agent runs automatically (requires write permission via /grant), Manual = use /agent |
| Agent API Timeout | 180000ms | Timeout per agent API call (auto-adjusted for complexity) |
| Agent Max Duration | 20 min | Maximum time for agent to run (5-60 min) |
| Agent Max Iterations | 50 | Maximum agent iterations (10-500). Modern models finish typical tasks in 3–8 iterations — raise for large autonomous refactors. |
| Agent Confirmation | Dangerous | Never, Dangerous (default), or Always |
| Agent Auto-Commit | Off | Automatically commit after agent completes |
| Agent Branch | Off | Create new branch for agent commits |
| Agent Auto-Verify | Off | Off, Build only, Typecheck only, Test only, or Build + Typecheck + Test |
| Agent Max Fix Attempts | 1 | Max attempts to auto-fix errors when Auto-Verify is enabled. More than 1 usually means the agent is stuck — bail and let the user decide. |
Add keys once on the dashboard, then sync them to any machine:
codeep account sync # Pull keys + config from codeep.dev → local
codeep account push # Push local keys + config → codeep.dev
Keys are encrypted at rest using AES-256-GCM.
Zed's left sidebar shows only sessions Zed itself created (one per "New Thread") — it does not auto-list sessions you saved from the CLI or from a previous Zed window. Two ways to access those:
Inside the chat (recommended):
/sessions # show all saved sessions for this workspace
/session load <name> # restore one
When you open a fresh chat in Zed, Codeep also shows you the most recent CLI sessions for that project right in the welcome message, so you can switch with one command.
Via Zed's import modal:
Open Zed's command palette (Cmd+Shift+P) and search for "Import Threads". Codeep's saved sessions appear there as an importable list — once imported they show up in Zed's regular sidebar.
/diff - Review unstaged changes with AI assistance/diff --staged - Review staged changes/commit - Generate conventional commit messages| Command | Description |
|---|---|
/diff | Review unstaged git changes |
/diff --staged | Review staged git changes |
/commit | Generate commit message for staged changes |
/git-commit [msg] | Commit current changes with message |
/push | Push to remote repository |
/pull | Pull from remote repository |
Codeep supports the Agent Client Protocol (ACP), letting you use it as an AI coding agent directly inside Zed.
@codeep chat participant — invoke Codeep from VS Code's built-in Chat view: type @codeep and ask, or @codeep /explain / @codeep /review with a selection. Answers come from your configured Codeep provider/model via the CLI (not VS Code's model picker), on an independent session.#codeepSkills language-model tool — exposes your workspace's Codeep skill bundles (.codeep/skills/*/SKILL.md) to VS Code agent mode and #-references, so the native aVia the agent settings panel you can toggle which tools ask for approval before running — same controls available in the TUI via /settings:
These apply in Dangerous confirmation mode (the default). Choose Always to confirm every action, or Never to skip all prompts.
Install Codeep from the VS Code marketplace. The extension is a thin UI around the CLI — all credentials and sessions live in the CLI's config (~/.config/codeep/config.json), so keys set anywhere are visible everywhere.
Open the Command Palette (Cmd+Shift+P / Ctrl+Shift+P) and run:
Codeep: Set API Key — pick a provider, paste the key, done. Writes into the CLI config, so running codeep in a terminal immediately sees the new key.Codeep: Open Chat — open the Codeep sidebar.Codeep: Review Current File — run /review <file> on the active editor.Codeep: New Session — start a fresh session in the chat.If you already set up keys in the CLI (or via the codeep.dev dashboard + codeep account sync), the extension picks them up automatically on first launch — no welcome prompt.
高潜力自动化AI工作流工具
AI Skill Hub 为第三方内容聚合平台,本页面信息基于公开数据整理,不对工具功能和质量作任何法律背书。
建议在沙箱或测试环境中充分验证后,再部署至生产环境,并做好必要的安全评估。
✅ Apache 2.0 — 宽松开源协议,可商用,需保留版权声明和 NOTICE 文件,含专利授权条款。
AI Skill Hub 点评:代码智能助手 的核心功能完整,质量优秀。对于自动化工程师和运维人员来说,这是一个值得纳入个人工具库的选择。建议先在非生产环境试用,再逐步推广。
| 原始名称 | Codeep |
| 原始描述 | 开源AI工作流:An autonomous AI agent that reads your whole project, runs commands, and writes 。⭐20 · TypeScript |
| Topics | aiai-agenttypescriptcli-app |
| GitHub | https://github.com/VladoIvankovic/Codeep |
| License | Apache-2.0 |
| 语言 | TypeScript |
收录时间:2026-06-09 · 更新时间:2026-06-09 · License:Apache-2.0 · AI Skill Hub 不对第三方内容的准确性作法律背书。
选择 Agent 类型,复制安装指令后粘贴到对应客户端