MCP Server

Add to Claude Code

claude mcp add searchlayerpdf \
  --env SLP_API_KEY=slp_live_xxxxxxxx \
  -- npx -y searchlayerpdf-mcp

Add to Claude Desktop

Add this to claude_desktop_config.json:

{
  "mcpServers": {
    "searchlayerpdf": {
      "command": "npx",
      "args": ["-y", "searchlayerpdf-mcp"],
      "env": { "SLP_API_KEY": "slp_live_xxxxxxxx" }
    }
  }
}

Get an API key from the dashboard. The server talks to the same REST API — your key scopes what it can do.

The eight tools

• analyze_pdfScore a PDF's pages without processing — see what would improve.
• process_pdfUpgrade a PDF and wait for the result. Takes optional quality (economy/quality) and privacy (open/private).
• get_job_statusCheck where a job is in the pipeline.
• get_job_resultFetch the download URL for a finished job.
• get_diagnosticsPer-page Retrievability Scores and engine routing.
• reprocess_jobRe-run a job, e.g. on a different tier.
• list_jobsList recent jobs on the account.
• delete_documentDelete a job and its stored documents.

Try it

Once connected, just ask:

"Use SearchLayerPDF to make contract-scan.pdf searchable, then tell me which pages it actually improved."

Claude will call process_pdf, poll with get_job_status, then read get_diagnostics — no glue code from you.