PlayCaller MCP Server

AI-powered sports intelligence for NFL, MLB, and NBA via the Model Context Protocol.

Connect Claude Desktop, Cursor, or any MCP-compatible client to live beat reporter signals, injury intelligence, cross-platform player identity resolution, anomaly detection, and Kalshi prediction market data.

Quick Start

1. Get a free API key

Start a 14-day free trial — no credit card required.

2. Add to Claude Desktop

Edit ~/.claude/claude_desktop_config.json (Mac/Linux) or %APPDATA%\Claude\claude_desktop_config.json (Windows):

{
  "mcpServers": {
    "playcaller": {
      "url": "https://playcallerapp.com/mcp",
      "headers": {
        "X-PlayCaller-Key": "your_api_key_here"
      }
    }
  }
}

Restart Claude Desktop. PlayCaller tools will appear automatically.

3. Verify with npx

npx @playcaller/playcaller-mcp

Prints the config block and tool list.

What You Get

| Data Type | Source | Refresh | |---|---|---| | Beat reporter injury signals | ESPN NFL, MLB, NBA | Every 2 hours | | Player identity resolution | ESPN, Sleeper, Yahoo, NFL.com | Daily | | Usage anomaly scores | 3-season baseline model | Daily | | Kalshi prediction market props | Kalshi API | 15-minute cache | | Fantasy point projections | PlayCaller scoring engine | On-demand |

13,606 players indexed across NFL, MLB, and NBA.

Available Tools

get_injury_signals

Active beat reporter intelligence signals. Filter by sport and severity.

sport     nfl | mlb | nba | all    (default: all)
severity  high | medium | low | all (default: all)
limit     1–50                      (default: 20)

Example prompt: "Get all high-severity NFL injury signals from PlayCaller."

get_player_identity

Resolve a player name or platform ID to a canonical PlayCaller identity with cross-platform mappings.

name         Player full name (fuzzy matched)
sport        nfl | mlb | nba
platform     espn | sleeper | yahoo | nfl.com
platform_id  Platform-specific player ID

Example prompt: "Look up Justin Jefferson's ESPN ID using PlayCaller."

get_anomaly_scores

Players whose current usage deviates from their 3-season baseline. Scores above 70 indicate a meaningful pattern break worth acting on.

sport      nfl | mlb | nba | all  (default: all)
min_score  0–100                  (default: 70)
limit      1–50                   (default: 10)

Example prompt: "Find NBA players with anomaly scores above 80 in PlayCaller."

get_prediction_markets

Active Kalshi prediction market props with PlayCaller AI confidence scores. A gap between confidence_score and market_probability indicates a potential edge.

sport     nfl | mlb | nba | all  (default: all)
min_edge  0–100 gap threshold
limit     1–30                   (default: 10)

Example prompt: "Show me MLB props where PlayCaller's confidence differs from the market by at least 15 points."

get_player_projections

AI-generated fantasy point projections for a player using the PlayCaller scoring engine. Accounts for active injury signals and recent usage trends.

player_name  required — full name
sport        required — nfl | mlb | nba
scoring      ppr | half_ppr | standard | points | 9cat

Example prompt: "Get PlayCaller's PPR projection for CeeDee Lamb."

Pricing

| Tier | Price | Rate Limit | Use Case | |---|---|---|---| | Free Trial | 14 days | Standard | Evaluate | | Hobbyist | $49/mo | 100 req/min | Personal tools, small apps | | Pro | $149/mo | 500 req/min | Production apps, APIs | | Enterprise | $1K–$5K/mo | Custom | Data partnerships |

All tiers access all 5 MCP tools. Get your key →

Protocol Details

• Transport: HTTP (JSON-RPC 2.0)
• Endpoint: POST https://playcallerapp.com/mcp
• Auth: X-PlayCaller-Key: <your_key> header
• Protocol version: 2024-11-05
• Discovery: GET https://playcallerapp.com/mcp returns tool list

Full API docs: playcallerapp.com/developer