经 AI Skill Hub 精选评估,hybridclaw 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 7.1 分,适合有一定技术背景的用户使用。
hybridclaw 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。
hybridclaw 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 hybridclaw # 方式二:npx 直接运行(无需安装) npx hybridclaw --help # 方式三:项目依赖安装 npm install hybridclaw # 方式四:从源码运行 git clone https://github.com/HybridAIOne/hybridclaw cd hybridclaw npm install npm start
# 命令行使用
hybridclaw --help
# 基本用法
hybridclaw [options] <input>
# Node.js 代码中使用
const hybridclaw = require('hybridclaw');
const result = await hybridclaw.run(options);
console.log(result);
# hybridclaw 配置说明 # 查看配置选项 hybridclaw --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 HYBRIDCLAW_CONFIG="/path/to/config.yml"
<img width="420" height="397" alt="HybridClaw - Enterprise-ready self-hosted AI assistant runtime" src="docs/hero.png" />
- Install from npm, source, or the multi-arch Nix flake; a preview Homebrew formula is available for --HEAD builds while stable tap publication is prepared. - hybridclaw gateway status reports sandbox/runtime details, and in container mode it includes the configured image name plus the resolved version and short image id. - hybridclaw backup creates a WAL-safe archive of the runtime home, and hybridclaw backup restore <archive.zip> validates the archive before replacing local runtime state. - hybridclaw update --yes upgrades a global npm install and auto-restarts a running local gateway with its original launch parameters when possible, falling back to hybridclaw gateway restart if not. - /admin/agents edits allowlisted bootstrap markdown files such as AGENTS.md, keeps saved revisions, and restores earlier versions from the browser. - /admin/statistics reports message, session, token, cost, and channel trends across a selected date range. - The Usage rollup surfaces loading skeletons, cost metrics, and per-model spend summaries without scanning every stored session on page load. - /admin/agent-scoreboard ranks agents by observed skill scores, reliability, timing, best skills, and CV links. - hybridclaw agent config accepts generated JSON payloads to upsert agent metadata, write bootstrap markdown, import profile images into the agent workspace, and optionally activate the agent. - /admin/channels edits transport config, encrypted channel credentials, Signal QR linking, Twilio voice settings, and per-channel instructions that are injected into prompts at runtime. - slack_webhook targets provide outbound-only Slack Incoming Webhook delivery with encrypted webhook URLs, named destinations, Block Kit text chunking, reachability status, and POST-only network policy grants. - discord_webhook targets provide outbound-only Discord Incoming Webhook delivery with encrypted webhook URLs, named destinations, message chunking, reachability status, and POST-only network policy grants. - /admin/approvals manages approval policies from the browser. - Approval policy evaluation runs through a hook-fed rule pipeline, so workspace policy ordering and plugin tool-use hooks share one approval path. - /admin/a2a-inbox shows read-only A2A message threads across the instance, with sender, recipient, timestamp, intent, and content for each message. - /admin/a2a-trust shows the local A2A public-key trust ledger for paired peer instances. - /admin/gateway reloads runtime config and refreshes secrets from the browser, and shows public URL plus tunnel status, without tearing down the enclosing workspace container; keep hybridclaw gateway restart for local/manual full restarts. - /context and the web chat context ring show current context-window usage, remaining headroom, and compaction counts for the active session. - /goal stores a standing completion condition for the current thread and queues supervised continuations until the goal is judged complete, paused, cleared, interrupted, or blocked by approval policy. - proactive.delegation.model can pin delegated work to a different model from the parent turn; /status shows delegate token totals and local-token share when that split is configured. - deployment.mode, deployment.public_url, deployment.tunnel.provider, and deployment.tunnel.health_check_interval_ms describe local/cloud exposure and tunnel health cadence. The built-in ngrok, Tailscale Funnel, and Cloudflare Tunnel providers read NGROK_AUTHTOKEN, TS_AUTHKEY, CLOUDFLARE_TUNNEL_TOKEN, and Cloudflare certificate credentials from the encrypted runtime secret store. - container.warmPool keeps a bounded adaptive pool of idle host/container runtimes for recently active agents when low cold-start latency matters. - container.persistBashState controls whether bash tool calls share shell state (cd, exported env vars, aliases) across turns in the same active runtime session; /admin/config exposes the same setting as Persistent bash state. - security.confidentialRedactionEnabled controls whether optional .confidential.yml rules redact prompts and block matching outbound text; /admin/config exposes the same setting as Confidential leak guard. - hybridclaw audit scan-leaks scans historical audit logs against optional .confidential.yml rules for NDA-class client, project, person, keyword, and regex matches. - Generated artifacts remain downloadable and attachable even when the sandbox exposes a custom workspace display root such as /app. - hybridclaw tui includes live delegate progress, pulsing tool rows, completion checkmarks, rendered Markdown tables, a keyboard-driven approval picker, and a ready-to-run hybridclaw tui --resume <sessionId> command on exit. Pressing Esc stops the active run and returns control to the prompt. - hybridclaw doctor checks runtime health including resource hygiene maintenance for stale gateway artifacts. hybridclaw doctor browser-use checks the local Playwright browser automation substrate and can install missing Chromium support with --fix. - hybridclaw onboarding and related local setup flows can restore the last known-good saved config snapshot or roll back to a tracked revision when config.json becomes invalid. - hybridclaw skill import supports community sources, local directories, and .zip archives. - hybridclaw skill install <source>, skill upgrade, skill revisions, and skill rollback manage packaged business skills with manifests, audit events, and snapshots. - hybridclaw skill list blocked and hybridclaw skill unblock <name> let local operators review scanner-blocked skills and record a bypass marker for the installed copy when the finding has been accepted. - The bundled tutorials cover owner, GTM, marketing, sales, DevRel, content, invoicing, webinar, and release-launch workflows that can run from the TUI, web chat, or connected channels. - hybridclaw eval hybridai-skills turns the bundled skills pages' "Try it yourself" prompts into a local eval suite, and live summaries surface the observed skill, artifact presence, and counted tool-call totals. - Channel delivery stays predictable: email seeds its first mailbox cursor from the current head instead of replaying old inbox mail, retry-aware transports honor server Retry-After backoff, expected transient Discord/Email/WhatsApp transport outages stay local with rate-limited logging, and WhatsApp startup avoids intermittent init-query bad-request failures.
| Capability | HybridClaw | OpenClaw | Hermes Agent |
|---|---|---|---|
| Self-hosted runtime | ✅ Gateway + sandboxed container runtime | ✅ Self-hosted gateway/runtime | ✅ Self-hosted gateway/runtime |
| Migration support | ✅ Imports from OpenClaw and Hermes | ❌ No comparable import path surfaced | ⚠️ Imports from OpenClaw only |
| Encrypted secrets | ✅ Encrypted store + SecretRefs | ⚠️ SecretRefs, not a built-in encrypted store | ⚠️ File-permission-based secret storage |
| Approvals / governance | ✅ Approvals, audit trails, sandbox, config history | ⚠️ Strong approvals/audit, less enterprise-governance framing | ⚠️ Strong approvals/isolation, less audit/admin surface |
| Memory / knowledge | ✅ Shared memory + HybridAI knowledge path | ⚠️ Strong memory/session features | ⚠️ Strong persistent/self-improving memory |
| Multi-agent observability | ✅ Built-in audit surfaces + platform path | ⚠️ Multi-agent/task inspection exists | ⚠️ Subagents + logs/session search, not central observability |
| Local + cloud deployment model | ✅ Local-first runtime with HybridAI cloud path plus SSH/Tailscale remote access | ⚠️ Self-hosted + remote access | ✅ Local, VPS, Docker, Modal, Daytona |
| Multiple UIs | ✅ TUI + Chat UI + Admin UI + Agents UI | ✅ TUI + WebChat + Control UI | ⚠️ TUI + messaging + API server, no comparable built-in admin/chat web UI |
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。
✅ MIT 协议 — 最宽松的开源协议之一,可自由商用、修改、分发,仅需保留版权声明。
AI Skill Hub 点评:hybridclaw 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 | hybridclaw |
| 原始描述 | 开源AI工作流:Enterprise-ready self-hosted AI assistant runtime with sandboxed execution, secu。⭐99 · TypeScript |
| Topics | workflowaiai-agentsanthropicchatgptclaude-codetypescript |
| GitHub | https://github.com/HybridAIOne/hybridclaw |
| License | MIT |
| 语言 | TypeScript |
收录时间:2026-05-22 · 更新时间:2026-05-22 · License:MIT · 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