MCP Server (Model Context Protocol)

Connect any AI coding assistant or autonomous agent directly to live Valorant esports data with zero setup, zero hosting, and no API keys required.

# Claude Code CLI (one command)
claude mcp add vlrdevapi -- uvx --from "vlrdevapi[mcp]" vlrdevapi-mcp

Or paste the drop-in JSON into your client config:

{
  "mcpServers": {
    "vlrdevapi": {
      "command": "uvx",
      "args": ["--from", "vlrdevapi[mcp]", "vlrdevapi-mcp"]
    }
  }
}

10-Second Setup by ClientCopy Anchor Link

Claude Code CLICopy Anchor Link

Run this single command in your terminal:

claude mcp add vlrdevapi -- uvx --from "vlrdevapi[mcp]" vlrdevapi-mcp

Claude Code will automatically install the server on demand, register all 40 tools, and make them available in your session.

Claude DesktopCopy Anchor Link

Open your configuration file:

• macOS: ~/Library/Application Support/Claude/claude_desktop_config.json
• Windows: %APPDATA%\Claude\claude_desktop_config.json
• Linux: ~/.config/Claude/claude_desktop_config.json

Add vlrdevapi under mcpServers:

{
  "mcpServers": {
    "vlrdevapi": {
      "command": "uvx",
      "args": ["--from", "vlrdevapi[mcp]", "vlrdevapi-mcp"]
    }
  }
}

Restart Claude Desktop. You will see the hammer icon showing 41 available Valorant tools.

Cursor IDECopy Anchor Link

1. Open Cursor Settings → Features → MCP Servers.
2. Click Add New MCP Server.
3. Name: vlrdevapi, Type: command, Command: uvx --from "vlrdevapi[mcp]" vlrdevapi-mcp.

Or add it directly to .cursor/mcp.json in your workspace:

{
  "mcpServers": {
    "vlrdevapi": {
      "command": "uvx",
      "args": ["--from", "vlrdevapi[mcp]", "vlrdevapi-mcp"]
    }
  }
}

Windsurf / CodeiumCopy Anchor Link

Add to ~/.codeium/windsurf/mcp_config.json:

{
  "mcpServers": {
    "vlrdevapi": {
      "command": "uvx",
      "args": ["--from", "vlrdevapi[mcp]", "vlrdevapi-mcp"]
    }
  }
}

Zed EditorCopy Anchor Link

In your Zed settings.json:

{
  "context_servers": {
    "vlrdevapi": {
      "command": "uvx",
      "args": ["--from", "vlrdevapi[mcp]", "vlrdevapi-mcp"]
    }
  }
}

Autonomous AI Agents (OpenCode, DeepSeek, Codex, Grok, etc.)Copy Anchor Link

For autonomous agents, CLI wrappers, or custom LLM frameworks, simply instruct your agent or set its tool runner to:

uvx --from "vlrdevapi[mcp]" vlrdevapi-mcp

Ready-to-Paste Agent PromptCopy Anchor Link

You can paste this system prompt or instruction into any agent:

Valorant Esports Data Integration: You have access to the VLRdevAPI Model Context Protocol (MCP) server. Run it via standard stdio using uvx --from "vlrdevapi[mcp]" vlrdevapi-mcp.

• Use composite tools like get_match_full_context(series_id) and get_team_overview(team_id) for high-level queries.
• Use matches_live() and matches_upcoming() for match schedules.
• Use player_agents(player_id) and get_player_career_summary(player_id) for player analysis.
• No API keys or external services are needed.

What You Can Ask Your AICopy Anchor Link

Once connected, your AI assistant can answer natural language prompts directly from live VLR.gg data:

• "Who is playing live right now in VCT?" → calls matches_live()
• "Show me the upcoming matches for Sentinels this week" → calls team_upcoming_matches(team_id=2593)
• "Compare FNATIC and LOUD head-to-head" → calls compare_teams(team_id_1=2593, team_id_2=6961)
• "Give me a career dossier on Aspas including top agents and prize money" → calls get_player_career_summary(player_id=8480)
• "What was the round-by-round economy and buy types in the Champions grand final?" → calls series_economy(series_id=754737)
• "Summarize the top Valorant esports news headlines" → calls get_news_digest()

High-Level Composite Tools (Best for AI)Copy Anchor Link

The server provides 9 composite tools designed specifically for multi-step AI reasoning. Instead of making 5 separate requests, your agent gets a complete dossier in one call:

Complete Granular Tool Catalog (40 Tools)Copy Anchor Link

In addition to composite tools, every single VLRdevAPI endpoint is exposed as an MCP tool:

• Matches: matches_upcoming, matches_live, matches_completed
• Series: series_info, series_players, series_rounds, series_performance, series_economy, series_replay, series_vods, series_map_summary
• Teams: team_info, team_roster, team_stats, team_completed_matches, team_upcoming_matches, team_transactions, team_placements
• Players: player_info, player_teams, player_agents, player_matches, player_profile, player_placements
• Events: events_list, event_info, event_stages, event_teams, event_matches, event_standings
• News: news_list, news_article

Why Zero Setup WorksCopy Anchor Link

1. No API Keys: VLR.gg data requires no authentication token or developer account.
2. uvx Isolation: uvx downloads vlrdevapi[mcp] to a temporary cache, executes the server, and keeps your system Python clean.
3. Graceful Error Trapping: Domain errors (NotFoundError, ValidationError) return structured JSON error payloads, preventing AI chat crashes.
4. Token-Efficient Serialization: Uses VLRdevAPI's optimized dictionary and JSON serialization to avoid bloating the model's context window.