TP

teoobarca/perplexity-mcp

Developer tools
14 stars 6 forks Quality 70 Trend 20MIT

Free Perplexity AI MCP server with multi-account pooling, React admin dashboard, zero-cost health monitoring, and Telegram alerts. No API keys needed.

Overview

Most Perplexity MCP servers are single-account wrappers around the paid Sonar API. - 🆓 — uses session cookies, not the paid API. Same features, zero per-query fees - 🏊 — round-robin across N accounts with automatic failover - 📊 — React UI to monitor quotas, manage tokens, tail logs in real-time - ❤️ — monitors all accounts via rate-limit API without consuming queries - 🛡️ — detects when Perplexity silently returns a regular result instead of deep research - 📱 — get notified when tokens expire or quota runs out - — fast, accurate answers with citations - — multi-model thinking for complex decisions - — comprehensive 10-30+ citation reports - — web, scholar, and social - sonar · gpt-5.2 · claude-4.5-sonnet · grok-4.1 - gpt-5.2-thinking · claude-4.5-sonnet-thinking - gemini-3.0-pro · kimi-k2-thinking · grok-4.1-reasoning - rotation across accounts - on failures (60s → 120s → ...

README


🎯 Why This One?

Most Perplexity MCP servers are single-account wrappers around the paid Sonar API. This one is different:

  • 🆓 No API costs — uses session cookies, not the paid API. Same features, zero per-query fees
  • 🏊 Multi-account pool — round-robin across N accounts with automatic failover
  • 📊 Admin dashboard — React UI to monitor quotas, manage tokens, tail logs in real-time
  • ❤️ Zero-cost health checks — monitors all accounts via rate-limit API without consuming queries
  • 🛡️ Downgrade protection — detects when Perplexity silently returns a regular result instead of deep research
  • 📱 Telegram alerts — get notified when tokens expire or quota runs out

✨ Features


🖼️ Screenshots


🚀 Quick Start

1. Clone & Install

git clone https://github.com/teoobarca/perplexity-mcp.git
cd perplexity-mcp
uv sync

2. Add to Your AI Tool

That’s it. Works immediately with anonymous sessions. Add your tokens for Pro access — see Authentication.


🛠️ Tools

Two MCP tools with LLM-optimized descriptions so your AI assistant picks the right one automatically:

perplexity_ask

AI-powered answer engine for tech questions, documentation lookups, and how-to guides.

Parameter Type Default Description
query string required Natural language question with context
model string null Model selection (see models)
sources array ["web"] Sources: web, scholar, social
language string en-US ISO 639 language code

Mode auto-detection: Models with thinking or reasoning in the name automatically switch to Reasoning mode.

"gpt-5.2"          → Pro Search
"gpt-5.2-thinking"  → Reasoning Mode  ← auto-detected

perplexity_research

Deep research agent for comprehensive analysis. Returns extensive reports with 10-30+ citations.

Parameter Type Default Description
query string required Detailed research question with full context
sources array ["web", "scholar"] Sources: web, scholar, social
language string en-US ISO 639 language code

[!TIP] Deep research takes 2-5 minutes per query. Provide detailed context and constraints for better results. The server has a 15-minute timeout to accommodate this.


🖥️ Admin Panel

A built-in web dashboard for managing your token pool. Start it with:

perplexity-server

Opens automatically at http://localhost:8123/admin/

Feature Description
📊 Stats Grid Total clients, Online/Exhausted counts, Monitor status
📋 Token Table Sortable columns, filter pills (Online/Exhausted/Offline/Unknown), icon actions
💰 Quota Column Per-token breakdown — Pro remaining, Research quota, Agentic research
❤️ Health Monitor Zero-cost checks via rate-limit API, configurable interval
📱 Telegram Alerts Notifications on token state changes (expired, exhausted, back online)
🔄 Fallback Toggle Enable/disable automatic Pro → free fallback
📥 Import/Export Bulk token management via JSON config files
📝 Log Viewer Live streaming, level filter (Error/Warning/Info/Debug), search, follow mode
🧪 Test Button Run health check on individual tokens or all at once

🔐 Authentication

By default, the server uses anonymous Perplexity sessions (rate limited). For Pro access, add your session tokens.

How to Get Tokens

  1. Sign in at perplexity.ai
  2. Open DevTools (F12) → Application → Cookies
  3. Copy these two cookies:
    • next-auth.csrf-token
    • next-auth.session-token

Single Token

Create token_pool_config.json in the project root:

{
  "tokens": [
    {
      "id": "my-account",
      "csrf_token": "your-csrf-token-here",
      "session_token": "your-session-token-here"
    }
  ]
}

Multi-Token Pool

Add multiple accounts for round-robin rotation with automatic failover:

{
  "monitor": {
    "enable": true,
    "interval": 6,
    "tg_bot_token": "optional-telegram-bot-token",
    "tg_chat_id": "optional-chat-id"
  },
  "fallback": {
    "fallback_to_auto": true
  },
  "tokens": [
    { "id": "account-1", "csrf_token": "...", "session_token": "..." },
    { "id": "account-2", "csrf_token": "...", "session_token": "..." },
    { "id": "account-3", "csrf_token": "...", "session_token": "..." }
  ]
}

[!NOTE] Session tokens last ~30 days. The monitor detects expired tokens and alerts you via Telegram.


⚙️ Configuration

Environment Variables

Variable Default Description
PERPLEXITY_TIMEOUT 900 Request timeout in seconds (15 min for deep research)
SOCKS_PROXY — SOCKS5 proxy URL (socks5://host:port)

Token States

Token state is computed automatically from session_valid + rate_limits (never set manually):

State Meaning Badge Behavior
🟢 normal Session valid, pro quota available Online Used for all requests
🟡 exhausted Session valid, pro quota = 0 Exhausted Skipped for Pro, used as auto fallback
🔴 offline Session invalid/expired Offline Not used for any requests
🔵 unknown Not yet checked Unknown Used normally (quota assumed available)

Fallback Chain

When a Pro request fails, the server tries progressively:

1. ✅ Next client with Pro quota (round-robin)
2. ✅ Next client with Pro quota ...
3. 🟡 Any available client (auto mode)
4. 🔵 Anonymous session (auto mode)
5. ❌ Error returned to caller

🏗️ Architecture

┌─────────────────────────────────────────────────────────┐
│  Your AI Assistant (Claude Code / Cursor / Windsurf)    │
└──────────────────────┬──────────────────────────────────┘
                       │ MCP (stdio)
                       ▼
┌──────────────────────────────────────────────────────────┐
│  perplexity-mcp                                          │
│  ┌────────────────┐  ┌────────────────────────────────┐  │
│  │  tools.py       │  │  server.py                     │  │
│  │  • ask          │──│  • Pool state sync             │  │
│  │  • research     │  │  • Timeout handling            │  │
│  └────────────────┘  └────────────────────────────────┘  │
└──────────────────────┬───────────────────────────────────┘
                       │
                       ▼
┌──────────────────────────────────────────────────────────┐
│  Backend Engine (perplexity/)                            │
│                                                          │
│  ┌─────────────┐  ┌──────────────┐  ┌────────────────┐  │
│  │  client.py   │  │  client_pool │  │  admin.py      │  │
│  │  • Search    │  │  • Rotation  │  │  • REST API    │  │
│  │  • Upload    │  │  • Backoff   │  │  • Static      │  │
│  │  • Validate  │  │  • Monitor   │  │    files       │  │
│  └──────┬───────┘  │  • Fallback  │  └────────┬───────┘  │
│         │          └──────────────┘           │          │
│         ▼                                     ▼          │
│  ┌─────────────┐                    ┌────────────────┐   │
│  │ Perplexity  │                    │ React Admin UI │   │
│  │ (web API)   │                    │ :8123/admin/   │   │
│  └─────────────┘                    └────────────────┘   │
└──────────────────────────────────────────────────────────┘
Component File Role
MCP Server src/server.py Stdio transport, pool state sync, timeout handling
Tool Definitions src/tools.py 2 MCP tools with LLM-optimized descriptions
API Client perplexity/client.py Perplexity API via curl_cffi (bypasses Cloudflare)
Client Pool perplexity/server/client_pool.py Round-robin, backoff, monitor, state persistence
Query Engine perplexity/server/app.py Rotation loop, 3-level fallback, validation
Admin API perplexity/server/admin.py REST endpoints + static file serving
Admin UI perplexity/server/web/ React + Vite + Tailwind dashboard

🧪 Development

# Install in development mode
uv pip install -e ".[dev]" --python .venv/bin/python

# Run unit tests (53 tests)
.venv/bin/python -m pytest tests/ -v

# Frontend development
cd perplexity/server/web
npm install
npm run dev      # Dev server with proxy to :8123
npm run build    # Production build

Project Structure

src/                          # MCP stdio server (thin wrapper)
  server.py                   #   Entry point, pool state sync
  tools.py                    #   Tool definitions

perplexity/                   # Backend engine
  client.py                   #   Perplexity API client (curl_cffi)
  config.py                   #   Constants, endpoints, model mappings
  exceptions.py               #   Custom exception hierarchy
  logger.py                   #   Centralized logging
  server/
    app.py                    #   Starlette app, query engine
    client_pool.py            #   ClientPool, rotation, monitor
    admin.py                  #   Admin REST API
    utils.py                  #   Validation helpers
    main.py                   #   HTTP server entry point
    web/                      #   React admin frontend (Vite + Tailwind)

tests/                        # 53 unit tests

⚠️ Limitations

  • Unofficial — uses Perplexity’s web interface, may break if they change it
  • Cookie-based auth — session tokens expire after ~30 days
  • Rate limits — anonymous sessions have strict query limits
  • Deep research — takes 2-5 minutes per query (this is normal)

📄 License

MIT

View the full README on GitHub

Install

This server does not publish a one-line install command.

Open the repository installation guide New to MCP? Start with the mcpskill.org guide →

Configuration

{ "mcpServers": { "perplexity": { "command": "uv", "args": ["--directory", "/path/to/perplexity-mcp", "run", "perplexity-mcp"] } } }