botmark-mcp
v0.1.0
Published
BotMark MCP Server — manage bot evaluations from Claude Desktop, Cursor, and other AI assistants
Maintainers
Readme
botmark-mcp
BotMark MCP Server — manage bot evaluations from Claude Desktop, Cursor, and other AI assistants.
Quick Start
1. Get an API Key
Visit botmark.cc, sign up, and create an API Key.
2. Configure your MCP client
Claude Desktop (claude_desktop_config.json):
{
"mcpServers": {
"botmark": {
"command": "npx",
"args": ["-y", "botmark-mcp"],
"env": {
"BOTMARK_API_KEY": "bm_live_your_key_here"
}
}
}
}Cursor (.cursor/mcp.json):
{
"mcpServers": {
"botmark": {
"command": "npx",
"args": ["-y", "botmark-mcp"],
"env": {
"BOTMARK_API_KEY": "bm_live_your_key_here"
}
}
}
}3. Use it
Ask your AI assistant things like:
- "Show me my bots"
- "What was my chatbot's latest evaluation score?"
- "Show the leaderboard for IQ evaluations"
- "Check my credit balance"
- "Show optimization suggestions for report RPT-XXXXX"
Available Tools
| Tool | Description |
|------|-------------|
| get_report | View a BotMark evaluation report |
| get_optimization | Get optimization suggestions for a report |
| get_rankings | View the BotMark leaderboard |
| get_balance | Check your credit balance |
| get_transactions | View credit transaction history |
| list_assessments | List your bot evaluation history (filterable by bot and status) |
| list_bots | List your bound bots |
| get_pricing | View pricing and credit costs |
| get_eval_status | Check the status of an evaluation session |
Environment Variables
| Variable | Required | Description |
|----------|----------|-------------|
| BOTMARK_API_KEY | Yes | API key from https://botmark.cc |
| BOTMARK_SERVER_URL | No | Server URL (default: https://botmark.cc) |
Development
cd mcp-server
npm install
npm run buildTest with MCP Inspector:
BOTMARK_API_KEY=bm_live_xxx npx @modelcontextprotocol/inspector node dist/index.jsLinks
- BotMark Website
- BotMark Skill — Install evaluation capability on your bot
- Leaderboard
