BK Jr. MCP Server
A hosted MCP server for the BK Jr. recruiter agent that integrates voice (Retell AI), SMS (Quo/OpenPhone), and Google services, exposing tools for candidate management, outreach, and scheduling.
README
BK Jr. MCP Server
A hosted MCP (Model Context Protocol) server for the BK Jr. recruiter agent. Wires together Retell AI (voice), Quo/OpenPhone (SMS), Google Sheets/Calendar/Gmail/Drive (via Composio).
Why this exists
The user-facing agent (BK Jr.) runs in Claude Desktop. Claude Desktop talks to the MCP server over HTTPS. This server:
- Adds bearer-token auth (the raw FastMCP server has none)
- Strips the trailing-slash redirect that breaks clients behind TLS-terminating proxies (Cloudflare, Fly, Render)
- Exposes a
/healthendpoint for the platform's health check - Runs in stateless mode (no long-lived sessions) so it survives restarts cleanly
Endpoints
| Path | Auth | Description |
|---|---|---|
GET /health |
none | Returns server status. Use for Render/Render health checks. |
POST /mcp |
Bearer | JSON-RPC: tools/list, tools/call, initialize, etc. |
POST /mcp/ |
Bearer | Same as /mcp (the trailing-slash alias). |
Auth
Every MCP request must include:
Authorization: Bearer <MCP_AUTH_TOKEN>
The server reads MCP_AUTH_TOKEN from env, falling back to SMS_AGENT_API_KEY for backward compat. If neither is set, the server refuses ALL MCP requests with 401.
Local development
docker build -t bk-jr-mcp:local .
docker run --rm -p 8090:8080 \
-e MCP_AUTH_TOKEN=test-token \
-e BACKEND_URL=http://host.docker.internal:8080 \
-e OPENPHONE_API_KEY=... \
-e RETELL_API_KEY=... \
-e COMPOSIO_API_KEY=... \
-e COMPOSIO_USER_ID=... \
-e COMPOSIO_CONNECTED_ACCOUNT_ID=... \
-e GOOGLE_SHEETS_ID=... \
bk-jr-mcp:local
Then test:
# No auth -> 401
curl -X POST http://localhost:8090/mcp
# With auth -> 200
curl -X POST http://localhost:8090/mcp \
-H "Authorization: Bearer test-token" \
-H "Content-Type: application/json" \
-d '{"jsonrpc":"2.0","id":1,"method":"tools/list"}'
Deploy to Render
This repo has a render.yaml blueprint. The simplest path:
- Push to GitHub
- Render dashboard → New → Blueprint → point at this repo
- Render reads
render.yamland creates the service - Set
MCP_AUTH_TOKENas a secret in Render - Set the Quo/Retell/Composio API keys as secrets too
- The service deploys and gets a public URL like
https://bk-jr-recruiter-mcp.onrender.com/mcp
Config
Required env vars (set as Render secrets):
MCP_AUTH_TOKEN— bearer token clients sendBACKEND_URL— the FastAPI backend (defaults tohttps://bk-jr-api.aixlabs.fun)OPENPHONE_API_KEY— Quo/OpenPhoneRETELL_API_KEY— Retell AICOMPOSIO_API_KEY(andCOMPOSIO_USER_ID,COMPOSIO_CONNECTED_ACCOUNT_ID,GOOGLE_SHEETS_ID)SMS_AGENT_API_KEY— bearer token for the backend's/api/*routes (fallback forMCP_AUTH_TOKEN)
24 tools exposed
| Domain | Tools |
|---|---|
| Candidates | list_candidates, get_candidate_by_phone, update_candidate |
| Jobs | list_jobs, get_job |
| SMS (Quo) | send_sms, list_phone_numbers, list_conversations |
| Voice (Retell) | trigger_screening_call, list_pending_screenings, list_recent_screenings |
| Retell agents | retell_list_agents, retell_get_agent, retell_create_agent, retell_place_call |
| Bulk | bulk_outreach, bulk_outreach_for_job |
| gmail_send, gcal_list_events, gcal_create_event, gdrive_create_folder, gdrive_share | |
| Control | pause_candidate, notify_bk |
Recommended Servers
playwright-mcp
A Model Context Protocol server that enables LLMs to interact with web pages through structured accessibility snapshots without requiring vision models or screenshots.
Magic Component Platform (MCP)
An AI-powered tool that generates modern UI components from natural language descriptions, integrating with popular IDEs to streamline UI development workflow.
Audiense Insights MCP Server
Enables interaction with Audiense Insights accounts via the Model Context Protocol, facilitating the extraction and analysis of marketing insights and audience data including demographics, behavior, and influencer engagement.
VeyraX MCP
Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.
graphlit-mcp-server
The Model Context Protocol (MCP) Server enables integration between MCP clients and the Graphlit service. Ingest anything from Slack to Gmail to podcast feeds, in addition to web crawling, into a Graphlit project - and then retrieve relevant contents from the MCP client.
Kagi MCP Server
An MCP server that integrates Kagi search capabilities with Claude AI, enabling Claude to perform real-time web searches when answering questions that require up-to-date information.
E2B
Using MCP to run code via e2b.
Neon Database
MCP server for interacting with Neon Management API and databases
Exa Search
A Model Context Protocol (MCP) server lets AI assistants like Claude use the Exa AI Search API for web searches. This setup allows AI models to get real-time web information in a safe and controlled way.
Qdrant Server
This repository is an example of how to create a MCP server for Qdrant, a vector search engine.