Connect an AI agent to Suited

Suited runs an MCP server, so any MCP-capable agent (Claude, Claude Code, Cursor, or one you build yourself) gets the same core abilities you have in the app: manage your career facts, generate a tailored and graded application for any job, and read your pipeline. Your agent finds the jobs and calls Suited to build the applications.

Endpoint

A stateless Streamable HTTP MCP server. Authenticate with a Suited API key.

POST  https://suitedforit.com/mcp
Authorization: Bearer sk_suited_your_key_here

Create and revoke keys inside Suited under Agents. You can copy a key there again anytime. Keys act on your account and only reach the MCP tools below, nothing else.

Drop-in skill

Beyond the raw tools, download the Suited skill - a portable instruction pack that teaches any agent the full apply workflow: check facts first, confirm before spending a generation, report grades honestly, hand over the rendered PDFs instead of rebuilding documents, and never submit without the user's approval.

Claude Code: unzip into ~/.claude/skills/. claude.ai / Claude Desktop: upload as a skill. Any other agent: paste the SKILL.md body into its instructions.

Connect in one line

claude.ai (web, desktop, mobile)

Settings → Connectors → Add custom connector. URL: https://suitedforit.com/mcp. Open Request headers and add the header Authorization with the value Bearer sk_suited_your_key_here. Save and enable it - Claude can now manage your facts and generate graded applications from any chat.

Claude Code

claude mcp add --transport http suited https://suitedforit.com/mcp \
  --header "Authorization: Bearer sk_suited_your_key_here"

Claude Desktop, Cursor, or any client (config file)

{
  "mcpServers": {
    "suited": {
      "url": "https://suitedforit.com/mcp",
      "headers": { "Authorization": "Bearer sk_suited_your_key_here" }
    }
  }
}

Tools

Tool	What it does
`list_facts`	Return your saved career facts (the raw material every application is tailored from).
`add_facts`	Append one or more atomic, truthful career facts. Input: `facts: [{ claim, category? }]`.
`generate_application`	Tailor a full application (CV, cover letter, and answers to any form questions) for a specific job and grade it with a mock screening panel. Input: `job_description`, optional `company`, `role`, `form_questions`, `include_letter`. Returns the graded package.
`list_applications`	Return your pipeline: each generated application with company, role, score, and status.
`get_application`	Return the full package (CV, cover letter, answers, grade, PDF links) for an application `id`.
`get_documents`	Download links for the rendered PDFs (`cv_pdf`, `letter_pdf`). Signed, work in any browser without auth, expire after 24h. Use these instead of rebuilding documents from JSON - the rendered PDFs carry the market-correct formatting and passed machine checks.

Billing: generate_application consumes one of your applications, exactly like generating in the app: your first 3 are free, then it needs an active subscription. Everything is written strictly from your real facts and graded honestly, never invented.

A typical agent loop

Point your agent at a job you like and let it do the rest:

1. list_facts        -> see what Suited knows about you
2. add_facts         -> fill any gaps the role needs
3. generate_application (job_description, company, role)
                     -> a tailored, graded CV + cover letter + answers
4. get_application (id) -> read it back, refine, apply