About sourcers-desk
sourcers-desk turns role descriptions into recruiter-style Boolean and x-ray search strings for recruiters, AI agents, and developers. Its output includes title variants, specialist skill terms, search strings, and iteration advice based on a private sourcing playbook.
Key features
- 01Expanded job-title variants
- 02Practitioner skill terms
- 03LinkedIn x-ray search strings
- 04Open-web resume search strings
- 05Search iteration advice
Use cases
- 01Build searches from a role description
- 02Find candidate profiles on LinkedIn
- 03Search the open web for resumes
MCP support
The service is available as an MCP connector for clients including Claude, ChatGPT, and Claude Code. Its tools return recruiting search outputs from a role description.
Configuration
Remote servers need no local install. Pick your client, then copy the config or install in one click.
https://sourcerdesk.com/mcpclaude mcp add --transport http com-sourcerdesk-sourcers-desk https://sourcerdesk.com/mcp{"mcpServers":{"com-sourcerdesk-sourcers-desk":{"type":"http","url":"https://sourcerdesk.com/mcp"}}}Open in Cursor{"servers":{"com-sourcerdesk-sourcers-desk":{"type":"http","url":"https://sourcerdesk.com/mcp"}}}Open in VS Code{"mcpServers":{"com-sourcerdesk-sourcers-desk":{"type":"http","url":"https://sourcerdesk.com/mcp"}}}{"mcpServers":{"com-sourcerdesk-sourcers-desk":{"serverUrl":"https://sourcerdesk.com/mcp"}}}Run commands in a terminal; paste JSON into the client’s MCP config and restart. Servers that need sign-in will prompt on first connect.
Tools
2| Parameter | Type | Required | Description |
|---|---|---|---|
| target | string | No | |
| location | string | No | |
| constraints | string | No | |
| role_description | string | Yes |
Details
From the project READMESourcer's Desk API — documentation
Pay-per-call recruiting expertise for AI agents and developers: POST a role description, get back production-grade boolean/x-ray search strings the way a 20-year recruiter actually writes them.
Live endpoint: POST https://sourcerdesk.com/api/v1/boolean
- What you get: expanded title variants (OR'd, capped where noise starts), practitioner "shibboleth" skill terms, LinkedIn x-ray strings with exact-phrase locations, open-web resume strings with area-code location groups, iteration advice.
- Pricing: 3 free calls/day/IP, then ~$0.02 per call via x402 on three rails — 0.05 XNO (feeless Nano), or $0.025 USDC on Base or Polygon.
- Quality: blind A/B-evaluated against a bare same-model LLM before launch — the playbook engine was preferred 10/10. Methodology and caveats: EVAL.md.
- For recruiters using Claude or ChatGPT: add it as an MCP connector in about 30 seconds, no wallet needed — MCP.md.
| Doc | Contents |
|---|---|
| API.md | Request/response schema, the 402 payment flow, curl and client examples |
| MCP.md | The MCP connector for Claude, ChatGPT, and Claude Code |
| EVAL.md | How the blind evaluation was run, the result, and its honest limits |
| examples/ | Real request, 402 quote, and response JSON |
What this repo is (and is not)
This is the public documentation. The playbook itself — the encoded judgment the API sells — is private and is not in this repository or any public one. The engine's outputs are yours to use freely; the rules that produce them are the product.
Links
- Product site: https://sourcerdesk.com
- Machine docs: https://sourcerdesk.com/api · https://sourcerdesk.com/llms.txt
- MCP registry:
com.sourcerdesk/sourcers-desk - Nano x402 client: https://github.com/Feeless402/feeless402
- Contact: [email protected]
Comments
New to MCP?
From what the protocol is to installing your first server, these answers cover most first questions.
An open protocol that defines how AI applications connect to external tools and data in a uniform way. Clients talk to the model; MCP servers provide tools, resources and prompt templates. They exchange standard messages, so one server can be reused by many clients.