[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"project-93189":3},{"id":4,"name":5,"fullName":6,"owner":7,"repo":5,"description":8,"homepage":9,"htmlUrl":10,"language":11,"languages":10,"totalLinesOfCode":10,"stars":12,"forks":13,"watchers":14,"openIssues":15,"contributorsCount":15,"subscribersCount":15,"size":15,"stars1d":15,"stars7d":15,"stars30d":16,"stars90d":15,"forks30d":15,"starsTrendScore":15,"compositeScore":17,"rankGlobal":10,"rankLanguage":10,"license":10,"archived":18,"fork":18,"defaultBranch":19,"hasWiki":20,"hasPages":18,"topics":21,"createdAt":10,"pushedAt":10,"updatedAt":29,"readmeContent":30,"aiSummary":31,"trendingCount":15,"starSnapshotCount":15,"syncStatus":32,"lastSyncTime":33,"discoverSource":34},93189,"ai-openclaw-cli","wikidjon\u002Fai-openclaw-cli","wikidjon","openfinclaw cli | ai openclaw mcp cli which works in claude code, cursor, and 20+ ai agents via mcp","",null,"TypeScript",135,696,51,0,102,55.53,false,"main",true,[22,23,24,25,26,27,28],"ai","cli","mcp-cli","openclaw","openclaw-mcp","openfinclaw","openfinclaw-mcp","2026-07-22 04:02:08","\u003Cdiv align=\"center\">\n\n**[English](README.md)** | **[中文](README.zh-CN.md)**\n\n\u003Cimg src=\"imgs\u002Flogo.svg\" alt=\"OpenFinClaw\" width=\"680\">\n\n### Your quant research team, in one prompt.\n\nResearch · strategy · backtest · paper trade — ship a complete quant workflow from a single natural-language prompt, inside Claude Code, Cursor, and 20+ AI agents.\n\n[![npm](https:\u002F\u002Fimg.shields.io\u002Fnpm\u002Fv\u002F@openfinclaw\u002Fcli)](https:\u002F\u002Fwww.npmjs.com\u002Fpackage\u002F@openfinclaw\u002Fcli) [![npm downloads](https:\u002F\u002Fimg.shields.io\u002Fnpm\u002Fdw\u002F@openfinclaw\u002Fcli)](https:\u002F\u002Fwww.npmjs.com\u002Fpackage\u002F@openfinclaw\u002Fcli) [![MCP compatible](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FMCP-compatible-8A2BE2)](https:\u002F\u002Fmodelcontextprotocol.io) [![License: MIT](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FLicense-MIT-blue.svg)](LICENSE)\n\n### 🚀 [Try it in 60 seconds — zero install](https:\u002F\u002Fhub.openfinclaw.ai\u002Fen\u002Fchat)\n\nRun a full research → strategy → backtest loop in your browser. No install, no API key, real market data.\n\n[Quick Start](#quick-start) · [Example Prompts](#example-prompts) · [Community](#community-leaderboard--fork--publish) · [Platforms](#supported-platforms) · [vs. other tools](COMPARISON.md)\n\n\u003C\u002Fdiv>\n\n---\n\n## What you get\n\n| | |\n|---|---|\n| 🧠 **DeepAgent analysis skills** | **60+** built-in — technical · fundamental · sentiment · risk · timing · factor |\n| 🌍 **Markets covered** | **5** — US equities · A-shares · HK · Crypto · Forex |\n| 🤖 **Works with** | **20+** AI platforms — Claude Code · Cursor · VS Code · Hermes · Windsurf · Codex … |\n| 🔄 **End-to-end flow** | research → strategy → backtest → paper trade → publish to the leaderboard |\n| ⚡ **How you interact** | streaming token-by-token in the terminal · MCP tool calls · in-browser playground |\n\n\u003Cp align=\"center\">\n  \u003Cimg src=\"imgs\u002Fdeepagent-backtest-metrics.png\" alt=\"DeepAgent backtest result — Tesla Bollinger Bands\" width=\"620\">\n  \u003Cbr\u002F>\n  \u003Csub>\u003Cem>Live output from \u003Ccode>openfinclaw deepagent research\u003C\u002Fcode> — one prompt: research → strategy → backtest → metrics.\u003C\u002Fem>\u003C\u002Fsub>\n\u003C\u002Fp>\n\n---\n\n## Example Prompts\n\nCopy-paste any of these into `openfinclaw deepagent research \"…\"` (or drop them straight into your AI agent). Each one runs the full research → strategy → backtest loop.\n\n**📈 Technical analysis**\n- `Find RSI divergence signals on NVDA in the last 6 months, then backtest them.`\n- `Compare a Bollinger Bands strategy on TSLA vs AAPL over 1 year — which wins?`\n- `Screen the S&P 500 for golden-cross signals this month.`\n\n**📊 Fundamentals & macro**\n- `Pull Apple's last 8 quarters of revenue, margins, and guidance. Summarize the trend.`\n- `What's driving the NVDA move this quarter — earnings, guidance, or narrative?`\n- `Compare AMD \u002F INTC \u002F NVDA on growth, margin, and valuation.`\n\n**🎯 Strategy generation**\n- `Design a momentum strategy on US mega-cap tech. Backtest 2y. Tell me where it breaks.`\n- `Write a mean-reversion strategy on BTC and show drawdown behavior through 2022.`\n- `A-shares 沪深 300 日内轮动策略，年化目标 15%，最大回撤 \u003C 10%。`\n\n**🧪 Backtest & stress-test**\n- `Backtest a 50\u002F200 SMA crossover on SPY from 2015. Include costs and slippage.`\n- `Stress-test my forked strategy against the 2020 and 2022 crashes.`\n\n> Want a ready-made one? Run `openfinclaw leaderboard` to browse the community's highest-ranked strategies, then `fork` any of them.\n\n---\n\n## Quick Start\n\n> 💡 Want to see it in action before installing? **[Try DeepAgent in your browser](https:\u002F\u002Fhub.openfinclaw.ai\u002Fen\u002Fchat)** first.\n\n### 60-second onboarding\n\n```bash\nnpx @openfinclaw\u002Fcli@latest install               # wizard + MCP configs + Skill registration + doctor\nopenfinclaw deepagent +research \"盘点 BTC 周线\"     # streaming research \u002F strategy \u002F backtest\n```\n\n`install` runs the interactive wizard, writes MCP configs to every detected AI agent, persists your `fch_` key to `~\u002F.openfinclaw\u002Fconfig.json` (chmod 600 on Unix), drops a `SKILL.md` under `~\u002F.claude\u002Fskills\u002Fopenfinclaw\u002F` so Claude Code \u002F Cursor auto-trigger on keywords like `quant` \u002F `backtest` \u002F `量化`, and finishes with a connectivity check.\n\nNon-interactive \u002F CI:\n\n```bash\nnpx @openfinclaw\u002Fcli@latest install --yes \\\n  --platforms cursor,claude-code --tool-groups deepagent,strategy \\\n  --api-key fch_xxx --register-skill\n```\n\nJust the wizard, no SKILL.md registration, no doctor: `npx @openfinclaw\u002Fcli init`.\n\n### CLI quick reference\n\nA single `fch_` key drives both DeepAgent and the strategy group. Resolution order: `--api-key` → `OPENFINCLAW_API_KEY` → `~\u002F.openfinclaw\u002Fconfig.json`.\n\n| Group | Commands |\n|-------|----------|\n| DeepAgent | `deepagent +research \"\u003Cquery>\"`, `deepagent health`, `deepagent skills`, `deepagent threads`, `deepagent messages`, `deepagent backtests`, `deepagent packages`, `deepagent download` |\n| Strategy | `leaderboard`, `strategy-info`, `fork`, `list-strategies`, `validate`, `publish`, `publish-verify` |\n| Raw | `api GET \u003Cpath>` · `api POST \u003Cpath> --json '\u003Cbody>'` — direct Hub Gateway call, auth pre-attached |\n| System | `install` · `init` · `skill-install` · `serve` · `doctor` · `update` · `examples` |\n\n`+verb` (e.g. `deepagent +research`) is the human-friendly streaming path; the bare verbs and MCP-only atomic triplet `research_submit \u002F research_poll \u002F research_finalize` are for agents\u002Fscripts. Run `openfinclaw --help` for the full surface.\n\n**Sample DeepAgent output** — one prompt → strategy definition + backtest metrics + per-trade P&L + improvement notes:\n\n\u003Cp align=\"center\">\n  \u003Cimg src=\"imgs\u002Fdeepagent-backtest-metrics.png\" alt=\"DeepAgent — strategy definition & performance metrics\" width=\"49%\" \u002F>\n  \u003Cimg src=\"imgs\u002Fdeepagent-backtest-trades.png\" alt=\"DeepAgent — trades, conclusions & optimization suggestions\" width=\"49%\" \u002F>\n\u003C\u002Fp>\n\n---\n\n## Community: leaderboard → fork → publish\n\nOpenFinClaw ships with a community strategy exchange. Browse what others are running, copy any strategy locally, tweak it, and publish back — think of it as a Hugging Face for quant strategies.\n\n```bash\nopenfinclaw leaderboard --limit 20          # Browse top-ranked strategies\nopenfinclaw strategy-info \u003Cid>              # See how a strategy performs\nopenfinclaw fork \u003Cid>                       # Copy to .\u002Fstrategies\u002F\u003Cslug>\n# ... edit strategy.py, tweak fep.yaml ...\nopenfinclaw validate .\u002Fstrategies\u002F\u003Cslug>    # Pre-flight FEP v2.0 check\nopenfinclaw publish .\u002Fmy-strategy.zip       # Ship to the leaderboard\nopenfinclaw publish-verify --submission-id \u003Cid>   # Track backtest progress\n```\n\nEvery published strategy is backtested server-side and ranked by live-market-equivalent returns — no self-reported numbers.\n\n---\n\n## Supported Platforms\n\nOpenFinClaw works with any MCP-compatible agent platform:\n\n| Category | Platforms |\n|----------|-----------|\n| **Chat** | Claude Desktop, Claude.ai, ChatGPT, Chatbox, LM Studio |\n| **IDEs** | Claude Code, VS Code (Copilot), Cursor, Windsurf, JetBrains Junie, Zed, Cline, Continue.dev |\n| **CLI Agents** | Codex (OpenAI), OpenCode, Amazon Q CLI |\n| **Frameworks** | Hermes Agent, BeeAI, Swarms |\n| **AI Agents** | OpenClaw, NanoClaw |\n| **Other** | v0 (Vercel), Postman, Roo Code, Amp (Sourcegraph) |\n\n### Platform Config Examples\n\n\u003Cdetails>\n\u003Csummary>\u003Cb>Claude Code\u003C\u002Fb> — \u003Ccode>~\u002F.claude\u002Fsettings.json\u003C\u002Fcode>\u003C\u002Fsummary>\n\n```json\n{\n  \"mcpServers\": {\n    \"openfinclaw\": {\n      \"command\": \"npx\",\n      \"args\": [\"@openfinclaw\u002Fcli\", \"serve\", \"--tools=deepagent,strategy\"],\n      \"env\": {\n        \"OPENFINCLAW_API_KEY\": \"fch_xxx\"\n      }\n    }\n  }\n}\n```\n\u003C\u002Fdetails>\n\n\u003Cdetails>\n\u003Csummary>\u003Cb>Cursor\u003C\u002Fb> — \u003Ccode>.cursor\u002Fmcp.json\u003C\u002Fcode>\u003C\u002Fsummary>\n\n```json\n{\n  \"mcpServers\": {\n    \"openfinclaw\": {\n      \"command\": \"npx\",\n      \"args\": [\"@openfinclaw\u002Fcli\", \"serve\", \"--tools=deepagent,strategy\"],\n      \"env\": {\n        \"OPENFINCLAW_API_KEY\": \"fch_xxx\"\n      }\n    }\n  }\n}\n```\n\u003C\u002Fdetails>\n\nFor other platforms (VS Code, Hermes, Windsurf, Zed, OpenClaw, Junie, Trae, …), see [`configs\u002F`](configs\u002F) for ready-to-copy templates. The shape is the same — only the host key (`servers` vs `mcpServers` vs `context_servers`) and config path differ.\n\n---\n\n## Tool Groups & Context Optimization\n\nLoad only what you need to save tokens: `serve --tools=deepagent` (~1,400 tk) or `serve --tools=strategy` (~1,000 tk), or omit `--tools` for both.\n\n| Group | Tools |\n|-------|-------|\n| `deepagent` | 14 remote-agent tools — `fin_deepagent_health` \u002F `_skills` \u002F `_research_submit` \u002F `_research_poll` \u002F `_research_finalize` \u002F `_status` \u002F `_cancel` \u002F `_threads` \u002F `_messages` \u002F `_backtests` \u002F `_backtest_result` \u002F `_packages` \u002F `_package_meta` \u002F `_download_package` |\n| `strategy` | 7 local FEP v2.0 tools — `strategy_publish` \u002F `strategy_validate` \u002F `strategy_fork` \u002F `strategy_leaderboard` \u002F `strategy_get_info` \u002F `strategy_list_local` \u002F `strategy_publish_verify` |\n\n---\n\n## Environment Variables\n\nOnly one is required:\n\n| Variable | Description |\n|----------|-------------|\n| `OPENFINCLAW_API_KEY` | Unified `fch_` key. Drives both strategy (Hub) and deepagent (Hub Gateway). Falls back to `~\u002F.openfinclaw\u002Fconfig.json` if unset. Get a key at [hub.openfinclaw.ai](https:\u002F\u002Fhub.openfinclaw.ai). |\n\nAdvanced overrides (rarely needed): `OPENFINCLAW_CONFIG_PATH`, `HUB_API_URL`, `DEEPAGENT_API_URL`, `REQUEST_TIMEOUT_MS`, `DEEPAGENT_SSE_TIMEOUT_MS` — see `packages\u002Fcore\u002Fsrc\u002Fconfig.ts`.\n\n---\n\n## Development\n\n```bash\ngit clone https:\u002F\u002Fgithub.com\u002Fmirror29\u002Fopenfinclaw-cli.git && cd openfinclaw-cli && pnpm install && pnpm build\nOPENFINCLAW_API_KEY=\u003Cfch_...> node packages\u002Fcli\u002Fdist\u002Findex.js doctor   # smoke test\n```\n\nMonorepo: `@openfinclaw\u002Fcore` (zero-dep business logic) + `@openfinclaw\u002Fcli` (MCP server + terminal CLI + install wizard).\n\n---\n\n## License\n\nMIT\n","这是一个面向量化金融研究的命令行工具，通过自然语言指令驱动端到端量化工作流（研究、策略生成、回测、模拟交易）。基于MCP（Model Context Protocol）协议，支持在Claude Code、Cursor等20+AI开发环境内调用，内置60+分析能力（技术\u002F基本面\u002F情绪\u002F风险等），覆盖美股、A股、港股、加密货币及外汇5类市场。适用于个人量化研究员、算法交易初学者及AI原生金融开发者快速验证策略想法，无需编码即可完成从提示到回测结果的完整闭环。",2,"2026-07-12 02:30:09","CREATED_QUERY"]