About Talent Summoner
Talent Summoner connects AI agents to candidate sourcing and review workflows for hiring teams. It supports candidate search, shortlisting, work-email lookups, expiring share links, and approval-gated outreach.
Key features
- 01Draft candidate profiles before searching
- 02Search and refine candidate criteria
- 03Review and shortlist candidates
- 04Request work-email lookups
- 05Share shortlists with expiring links
- 06Prepare outreach for approval
Use cases
- 01Source candidates for an open role
- 02Review and shortlist candidates with an AI agent
- 03Share a shortlist through a temporary link
- 04Approve outreach after reviewing recipients and messages
MCP support
Talent Summoner exposes sourcing tools through a remote MCP server at `https://talentsummoner.com/api/mcp`. It supports Streamable HTTP, OAuth 2.1 with PKCE, or a Bearer-token API key.
Configuration
Remote servers need no local install. Pick your client, then copy the config or install in one click.
https://talentsummoner.com/api/mcpclaude mcp add --transport http com-talentsummoner-talent-summoner https://talentsummoner.com/api/mcp{"mcpServers":{"com-talentsummoner-talent-summoner":{"type":"http","url":"https://talentsummoner.com/api/mcp"}}}Open in Cursor{"servers":{"com-talentsummoner-talent-summoner":{"type":"http","url":"https://talentsummoner.com/api/mcp"}}}Open in VS Code{"mcpServers":{"com-talentsummoner-talent-summoner":{"type":"http","url":"https://talentsummoner.com/api/mcp"}}}{"mcpServers":{"com-talentsummoner-talent-summoner":{"serverUrl":"https://talentsummoner.com/api/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
Details
From the project READMETalent Summoner for AI agents
Connect your AI agent to Talent Summoner to source and review hiring candidates. The Model Context Protocol (MCP) connection gives your AI agent sourcing tools, and the companion skill guides it through the workflow.
Works with MCP-compatible AI agents, including Claude, ChatGPT and Cursor. Full connection guide, tool list and privacy details: talentsummoner.com/mcp.
Quick start
-
Sign in to Talent Summoner and create an API key.
-
In an interactive macOS or Linux terminal, or Windows PowerShell or Command Prompt, with Node.js 22.20 or newer, run:
npx @talent-summoner/setup -
Select your AI agents, paste the key into the hidden terminal prompt, and let setup check the connection. It installs the MCP connection and skill for the AI agents you select.
-
Open or restart each AI agent, approve the connection if prompted, and ask: “Show my Talent Summoner sourcing sessions.”
Keep your API key out of AI agent chats, shell commands, and Git. Setup itself does not start sourcing or use a Role.
Install as a Claude Code plugin
The plugin bundles the MCP connection and the skill, and you sign in with your Talent Summoner account instead of an API key.
In Claude Code:
/plugin marketplace add talent-summoner/agent-skills
/plugin install talent-summoner@talent-summoner
Claude Code asks you to sign in to Talent Summoner the first time it uses a tool. The plugin runs no local code, hooks or scripts; it only connects to https://talentsummoner.com/api/mcp. Disconnect at any time in Talent Summoner under Settings → Connected apps. Use either the plugin or npx @talent-summoner/setup in Claude Code, not both, to avoid duplicate tools and skills. For Cursor and other AI agents, use npx @talent-summoner/setup.
ChatGPT and Codex plugin
We have prepared a Talent Summoner plugin for the OpenAI plugin directory used by ChatGPT and Codex. It is not listed yet and will be available there only after OpenAI approves it. Until then, use npx @talent-summoner/setup for Codex, or add the MCP connection manually.
What you can do
- Draft an ideal candidate profile before searching: “Draft the ideal candidate profile for a senior React engineer in Berlin. Let me review it before searching.”
- Search for candidates, refine the criteria, and find more candidates in the same session.
- Review and shortlist candidates, and request work-email lookups.
- Share a shortlist through a link that expires.
- Prepare outreach to shortlisted candidates and send it once you approve.
Your AI agent asks before unlocking a Role for a session. It also asks before publishing a share link. Before any outreach goes out, it shows you the sender, every recipient and every complete message, and sends only after you approve. Sending cannot be undone.
Manage your connection
Rerun npx @talent-summoner/setup to replace a key or update the skill. To disconnect, revoke the key, then remove the talent-summoner MCP entry and skill from your AI agent. Setup displays their locations during installation.
If your AI agent does not show the tools after setup, restart it and check its MCP connection or workspace approval. Report a setup issue if the connection still fails.
Install the skill separately
If you have configured the Talent Summoner MCP connection yourself, install only the companion skill:
npx skills add talent-summoner/agent-skills -g
This command installs the skill instructions. It does not set up the MCP connection. Use one skill installation method per AI agent to avoid duplicate entries.
For a manual MCP connection, add an HTTP server at https://talentsummoner.com/api/mcp in your AI agent's MCP settings and authenticate with your API key as a Bearer token.
Links
- Talent Summoner MCP page: connection guide, tools and FAQ
- Remote MCP endpoint: Streamable HTTP, OAuth 2.1 with PKCE, or API key as a Bearer token
- Privacy policy and terms
- Report an issue
Contributing
See the contributor guide for local development and releases.
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.