经 AI Skill Hub 精选评估,Rust Agent Development Kit (ADK-Rust) 获评「推荐使用」。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 7.5 分,适合有一定技术背景的用户使用。
Rust Agent Development Kit (ADK-Rust) 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。
Rust Agent Development Kit (ADK-Rust) 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:cargo install(推荐) cargo install adk-rust # 方式二:从源码编译 git clone https://github.com/zavora-ai/adk-rust cd adk-rust cargo build --release # 二进制在 ./target/release/adk-rust
# 查看帮助 adk-rust --help # 基本运行 adk-rust [options] <input> # 详细使用说明请查阅文档 # https://github.com/zavora-ai/adk-rust
# adk-rust 配置说明 # 查看配置选项 adk-rust --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 ADK_RUST_CONFIG="/path/to/config.yml"
A production-ready Rust framework for building AI agents. Model-agnostic, type-safe and async, across 43 publishable crates for agent orchestration.
v2.2.0 Released! This API-compatible minor release completes the Gemini Enterprise Agent Platform consumption path: the Gen AI Evaluation Service bridge, Vertex AI RAG Engine retrieval and grounding, an Agent Retrieval vector store, Agent Registry discovery and registration, Skill Registry consumption with remote skill loading, and remote ReasoningEngine agents you can call as sub-agents — every one opt-in and composable with any preset, and all appended to gemini-agent-platform. Graph workflows gain native tool confirmation pauses. Tracing is fixed so one invocation exports as one trace rather than several disconnected ones. All 43 crates are available on crates.io. Milestone: ADK-Rust has crossed 500K total crates.io downloads across the workspace crates. Coming from 1.x: six APIs changed shape and the fan-in default changed behaviour without an API change. See the migration guide and the CHANGELOG.
```
| Tier | Includes | Use case |
|---|---|---|
minimal (default) | Gemini provider, agents, runner, sessions | Fast starter agents |
standard | minimal + OpenAI, Anthropic, tools, memory, telemetry, server, auth, graph, eval, guardrail, plugins, artifacts, skills | Serving an agent over HTTP |
enterprise | standard + realtime, browser, RAG, payments, AWP | Voice, retrieval and payments |
full | enterprise + audio, code execution, sandbox | Everything |
A tier is a starting point, not a ceiling. Add any single capability on top of one without moving to the next tier, so features = ["minimal", "audio"] gives you the minimal build plus audio. AGENTS.md lists every feature you can add this way.
Scaffold an OpenAI agent with the HTTP runtime and embedded UI:
```bash cargo install cargo-adk cargo adk new quickstart_agent --template api --provider openai cd quickstart_agent cp .env.example .env
Each row links to its guide and a runnable example.
| Capability | Guide | Example |
|---|---|---|
| Embedded runtime UI — conversations, Markdown, tools, workflow/team topology, realtime playback, protocols, state and telemetry | [deployment](docs/official_docs/deployment/server.md#web-ui) | [examples/advanced_agents](examples/advanced_agents) |
Tools — #[tool] derives the JSON schema from your argument type | [tools](docs/official_docs/tools/function-tools.md) | [examples/coding_agent](examples/coding_agent) |
MCP clients and servers on rmcp 3.1 — tools, resources, prompts, elicitation, tasks | [mcp](docs/official_docs/mcp/index.md) | [examples/mcp_protocol_revisions](examples/mcp_protocol_revisions) |
| Workflow agents — sequential, parallel, loop | [agents](docs/official_docs/agents/workflow-agents.md) | [examples/multi_perspective_analysis](examples/multi_perspective_analysis) |
| Portable teams — validated handoff, delegation, policies, receipts and shared state | [multi-agent](docs/official_docs/agents/multi-agent.md) | [examples/team_architectures](examples/team_architectures) |
| Graph workflows — checkpoints, durable resume, human-in-the-loop, subgraphs | [graph-agents](docs/official_docs/agents/graph-agents.md) | [examples/graph_subgraph_claims](examples/graph_subgraph_claims) |
| Coding agents — read, edit and run code in a confined workspace | [coding-agent](docs/official_docs/coding-agent/index.md) | [examples/coding_goal](examples/coding_goal) |
| Realtime voice and video — OpenAI Realtime, Gemini Live, Vertex, LiveKit, WebRTC | [realtime](docs/official_docs/agents/realtime-agents.md) | [examples/realtime_voice](examples/realtime_voice) |
| Governed computer use — approval interrupts bound to a digest | [computer-use](docs/official_docs/computer-use/index.md) | — |
| RAG — chunking, embeddings, vector search, 6 backends | [rag](docs/official_docs/tools/rag.md) | — |
| Memory — semantic search, project isolation, a bi-temporal knowledge graph | [memory](docs/official_docs/tools/memory-tools.md) | [examples/skill_memory_improvements](examples/skill_memory_improvements) |
| Servers — REST with SSE, A2A v1.0.0, background runs, cron | [deployment](docs/official_docs/deployment/server.md) | [examples/ambient_cron_agent](examples/ambient_cron_agent) |
| Gemini Enterprise Agent Platform — Agent Engine BYOC, managed state, memory, artifacts, telemetry and sandbox | [agent-engine](docs/official_docs/deployment/agent-engine.md) | [examples/vertex_sandbox](examples/vertex_sandbox) |
| Agentic Web Protocol — discovery, manifests, trust levels, consent | [awp](docs/official_docs/deployment/awp.md) | [examples/awp_agent](examples/awp_agent) |
| Agentic commerce — ACP and AP2 with durable journals | [payments](docs/official_docs/security/payments.md) | [examples/payments](examples/payments) |
| Editor interop — use an ACP coding agent as a tool, or expose yours | [acp](docs/official_docs/acp/index.md) | — |
| Browser automation — 46 WebDriver tools | [browser-tools](docs/official_docs/tools/browser-tools.md) | — |
| Evaluation — trajectory, rubric, LLM-judge, A/B, CI output | [evaluation](docs/official_docs/evaluation/evaluation.md) | [examples/eval_showcase](examples/eval_showcase) |
| Guardrails, RBAC, SSO, audit logging | [security](docs/official_docs/security/access-control.md) | — |
| Observability — OpenTelemetry tracing, structured logging | [observability](docs/official_docs/observability/telemetry.md) | [examples/advanced_agents](examples/advanced_agents) |
cargo run
Open [http://127.0.0.1:8080/ui/](http://127.0.0.1:8080/ui/), enter a prompt,
and press <kbd>Enter</kbd>. The UI creates the session, streams the run, renders
Markdown and tool results, animates the active agent or workflow edge, and keeps
the event timeline, state, artifacts, and telemetry beside the conversation.

The animation uses the richer team showcase so the topology is visible; the
single-agent project you just generated uses the same UI with a one-node graph.
Confirm the server independently with:
bash curl -fsS http://127.0.0.1:8080/api/health ```
The five-minute quickstart explains the generated files and the console-only alternative. The runnable runtime_ui_showcase reproduces the UI above with tool, graph, and team agents.
ADK-Rust 是一个专为 Rust 开发者设计的全功能 AI Agent 构建框架。它提供了一套高度可组合的 Template System,包含 8 个基础模板、9 个插件及 5 种企业级模式,支持通过 `cargo adk new --addon` 命令快速进行项目脚手架搭建。此外,框架内置了 `cargo adk build` 工具,允许开发者在不进行实际部署的情况下编译并验证 Agent 项目,从而在 CI/CD 和本地开发中实现极速的反馈循环,并提供类型安全的 Agent 抽象。
ADK-Rust 具备强大的生产级特性:在会话管理方面,支持基于内存或 SQLite 的 Session Management,并提供 AES-256-GCM 加密及密钥轮转功能;在记忆系统方面,支持具备语义搜索和向量嵌入(Vector Embeddings)能力的长期记忆,并实现项目级的隔离;在通信方面,提供支持 SSE 流式传输的 REST API,以及用于 Agent 间通信的 A2A v1.0.0 协议,可通过 `A2aServer::quick_start` 快速启动。
使用本项目需要满足以下环境要求:Rust 版本需为 1.85 或更高(支持 Rust 2024 edition)。若需使用 Vertex AI Live 功能,需配置 gcloud auth application-default;若使用 LiveKit Bridge,需部署 LiveKit server;若需使用 OpenAI WebRTC 功能,则系统需安装 cmake。开发者需根据所选的特性(如 vertex-live, livekit, openai-webrtc)配置相应的 feature flags。
可以通过两种方式安装 ADK-Rust:1. 直接集成到现有项目,在 `Cargo.toml` 中添加 `adk-rust = { version = "0.9.1", features = ["standard"] }`,这是包含 Gemini 模型、Agent 运行时及 Session 管理的最小化配置;2. 从源码构建,适用于需要自定义修改或深度开发的场景。建议在开发前通过 `devenv shell` 或 Nix 环境进行初始化,以确保开发环境的可复现性。
框架提供了简洁的快速入门示例。开发者可以通过 `LlmAgentBuilder` 轻松构建基于 Gemini 或 OpenAI 的 Agent。例如,在集成 OpenAI 时,只需在 `Cargo.toml` 中开启 `openai` feature,即可通过 `OpenAIClient` 初始化模型,并结合 `Launcher` 运行 Agent。代码逻辑清晰,支持通过环境变量(如 `OPENAI_API_KEY`)安全地管理 API 密钥,非常适合快速原型开发与生产部署。
项目的配置管理非常灵活。对于生产环境,建议通过 `aws configure` 或环境变量进行配置。为了确保在 Linux 和 macOS 等不同操作系统上拥有完全一致的开发体验,推荐使用 Nix 或 devenv 进行环境管理(通过 `devenv shell` 命令)。这种方式可以实现高度的可复现性,避免因本地环境差异导致的配置问题。
ADK-Rust 的 API 设计面向生产级应用,特别针对非 CLI 提供商分发(fan-out)的场景进行了优化。通过在 `Cargo.toml` 中引入 `adk-rust = { version = "0.9.1", features = ["standard"] }`,开发者可以获得一套标准化的 API 接口,用于构建高性能、高可靠性的 AI 服务,确保在复杂的生产环境中依然保持稳定的响应能力。
对于需要构建复杂、有状态逻辑的开发者,ADK-Rust ��供了基于图的工作流能力。通过 `adk-graph` crate,你可以实现类似 LangGraph 的开发体验,利用 `AgentNode` 构建复杂的任务编排。你可以定义不同的 LLM Agents(如翻译 Agent)并将其连接成有向图,从而实现精细化的任务流转与状态管理,非常适合处理复杂的 AI 业务逻辑。
该项目提供了一个开源的AI工作流解决方案,使用Rust语言构建模块化的AI代理,支持Google服务集成,值得关注。
该工具使用 NOASSERTION 协议,商用场景请仔细阅读协议条款,必要时咨询法律意见。
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。
📄 NOASSERTION — 请查阅原始协议条款了解具体使用限制。
AI Skill Hub 点评:Rust Agent Development Kit (ADK-Rust) 的核心功能完整,质量良好。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 | adk-rust |
| 原始描述 | 开源AI工作流:Rust Agent Development Kit (ADK-Rust): Build AI agents in Rust with modular comp。⭐347 · Rust |
| Topics | workflowadkadk-agentadk-artifactadk-cliadk-googlerust |
| GitHub | https://github.com/zavora-ai/adk-rust |
| License | NOASSERTION |
| 语言 | Rust |
收录时间:2026-05-24 · 更新时间:2026-05-30 · License:NOASSERTION · 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