PolyCLIDash

A browser-based dashboard that wraps the Polymarket CLI with live data feeds, one-click trading, Gamma API market search, and AI-powered (GLM-5 / any OpenAI-compatible model) trade research.

Community: discord.gg/RSDX3PbCyD

---

Features

Feature	Description
🔍 Market Search	Live search via the Polymarket Gamma API — no market ID needed upfront
📡 Live Feed	Auto-refreshes markets, positions, open orders, and orderbook on a timer
⚡ Quick Trade	Buy / sell form with confirmation dialog before any order is placed
🛠 Action Buttons	Click-friendly wrapper for all your CLI presets
🤖 AI Research	Structured trade analysis from GLM-5 (or any OpenAI-compatible model) in markdown
🔒 Security	Helmet CSP, rate limiting (120/min general, 20/min heavy endpoints)
🖥 UX	Toast notifications, copy-to-clipboard, one-click "Use this market" fill, markdown rendering

---

Quick Start

1 — Prerequisites

• Node.js 20+
• Polymarket CLI installed and authenticated
• Ollama with glm-5 pulled or an OpenAI-compatible API key

2 — Install

git clone https://github.com/krutftw/PolycliDash
cd PolycliDash
npm install

3 — Configure

cp .env.example .env
# edit .env with your values

4 — Run

npm start

Open http://localhost:3000 in your browser.

---

Configuration (.env)

Variable	Default	Description
PORT	3000	HTTP server port
CLI_PATH	polymarket	Path to the Polymarket CLI binary
CLI_TIMEOUT_MS	30000	Max time (ms) to wait for a CLI command
OLLAMA_BASE_URL	http://localhost:11434	Ollama API endpoint
LIVE_REFRESH_INTERVAL_MS	9000	How often the live feed auto-refreshes (ms)

---

Market Search

The Market Search panel uses the Polymarket Gamma API (gamma-api.polymarket.com/markets).

• Leave the search box blank to load trending markets
• Type any keyword (e.g. "Bitcoin", "election", "Fed rate") and press Search
• Each result card shows live YES/NO outcome prices
• Click → Use Market ID to fill the Market ID into every panel at once
• Click → Use Token ID on a specific outcome to fill the Token ID

---

AI Research

1. Use Market Search or manually enter a Market ID
2. Type your question (e.g. "Which side is stronger right now?")
3. Choose time horizon and risk style
4. Click Get Research

The AI receives: live market data, orderbook depth, your positions, current market-implied probabilities, and a structured analyst system prompt. Output is rendered in markdown.

Custom AI model

Expand Custom AI Model to:

• Switch from Ollama to any OpenAI-compatible API
• Override the model name and base URL
• Enter an API key if required

---

Development

npm test        # run all 15 unit tests (Vitest)
npm run lint    # ESLint (if configured)

Tests are in src/server/__tests__/.

---

Architecture

src/
  server/
    app.js              — Express app factory (routes, middleware)
    index.js            — HTTP server + graceful shutdown
    config.js           — Env-based config
    cliRunner.js        — Spawns CLI subprocesses
    liveOverview.js     — Aggregates live CLI data
    gammaClient.js      — Polymarket Gamma API client
    ollamaClient.js     — Ollama AI client (GLM-5 enforced)
    openAiCompatibleClient.js — OpenAI-compatible AI client
    researchPrompt.js   — Builds structured AI research messages
    setupWizard.js      — Checks CLI/API/auth/presets
    aiConfig.js         — Validates AI provider config
    presets.js          — Loads CLI action presets
  server/__tests__/     — Vitest unit tests

public/
  index.html            — Dashboard HTML
  app.js                — Frontend logic (vanilla JS ES modules)
  styles.css            — All styles (dark matrix theme)

config/
  cli-presets.json      — Preset CLI actions shown in Action Buttons

---

Security

• Helmet — sets secure HTTP headers including a strict Content Security Policy
• Rate limiting — 120 req/min on all /api routes; 20 req/min on /api/research and /api/setup/wizard
• Graceful shutdown — SIGTERM/SIGINT handled cleanly with an 8-second force-exit timeout
• No secrets stored client-side; AI API keys stay in memory only for the request lifetime

---

License

MIT