Integration

flatmark integration

flatmark converts documents to Markdown: PDF, Word (.docx), PowerPoint (.pptx), Excel (.xlsx), HTML and plain text. Convert a file up to 8 MB in one call (1 credit), or queue a file up to 25 MB and 200 pages for OCR, table and heading recognition (10 credits, refunded on failure) and download its Markdown or its JSON structure; check plan, rate limit and credits, and manage webhook endpoints for job events. flatmark does not fetch URLs: send the file itself. Authentication is an API key in the X-API-Key header, created after signing in at flatmark.dev; without a key the direct conversion runs at a lower per-IP limit.

Connect flatmark How integrations work

Ask it in chat

Say what should happen.
The workflow builds itself.

When this happens, do that in flatmark.Built from the 11 flatmark tools below, on the connection you add once.

What you can do with flatmark

Each of these is one operation on one endpoint, available both as a workflow step and as a tool an AI agent can call on its own. Nothing bundles several operations into one action, so what a run did stays readable afterwards.

Each badge shows the default agent permission of an endpoint (allowed: an agent calls it without asking; ask first: you are asked before an agent calls it, when the agent or chat asks permission for sensitive actions; never: no agent may call it), which every user can change endpoint by endpoint in their credential settings.

Most endpoints run on the free plan. A few, mainly the ones that publish on your behalf, need a paid plan.

Job, 3 operations

get_jobGet jobGETAgents: ask firstReturns one queued job of the account: status (queued, running, succeeded or failed), error and timestamps. An unknown id, or another account's job, answers 404. Free.
list_jobsList jobsGETAgents: ask firstLists the account's queued conversion jobs, newest first, with status (queued, running, succeeded, failed), error and timestamps. Pass next_cursor back as cursor for the next page. Free.
submit_conversion_jobSubmit conversion jobPOSTAgents: ask firstQueues a file (PDF, DOCX, PPTX, XLSX, HTML, text; 25 MB, 200 pages) for OCR and tables; 10 credits, refunded on failure. Read the Markdown with get_conversion_result. Waits for the job and returns the finished result; gives up after 5 minutes.

Webhook, 3 operations

create_webhook_endpointCreate webhook endpointPOSTAgents: ask firstSubscribes an https URL to job events (such as job.completed and job.failed); flatmark then POSTs each event there, signed per Standard Webhooks. At most 10 endpoints per account. The signing secret is not returned by this tool.
delete_webhook_endpointDelete webhook endpointDELETEAgents: ask firstDeletes a webhook endpoint and its delivery history; pending deliveries to it are not sent. This cannot be undone. Free.
list_webhook_endpointsList webhook endpointsGETAgents: ask firstLists the account's webhook endpoints (at most 10), oldest first: id, url, subscribed event types, enabled flag and dates. Signing secrets are never returned. Free.

Conversion result, 2 operations

get_conversion_resultGet conversion resultGETAgents: ask firstReturns the Markdown of a finished queued job (job_id from submit_conversion_job or list_jobs) in markdown. Free. Fails with job_not_ready while it runs, job_failed if it failed, result_expired after 7 days.
get_conversion_result_jsonGet conversion result jsonGETAgents: ask firstReturns the DoclingDocument JSON of a finished queued job (job_id from submit_conversion_job): texts, tables and pictures with labels and page numbers. Much larger than the Markdown. Free.

Account usage, 1 operation

get_account_usageGet account usageGETAgents: ask firstReturns the plan the call runs on (tier), whether the key was recognized, the requests left in the current rate-limit window, the plan's features and, on credit plans, the credits left. Free; does not count against the rate limit.

Convert, 1 operation

convert_documentConvert documentPOSTAgents: ask firstConverts one file (PDF, DOCX, PPTX, XLSX, HTML or plain text, at most 8 MB) to Markdown and returns it in markdown; 1 credit per call. PDFs come back as plain text without headings or OCR: for scans or tables use submit_conversion_job.

Webhook test event, 1 operation

send_webhook_test_eventSend webhook test eventPOSTAgents: ask firstSends a real webhook.test event to the endpoint's URL within seconds, whatever event types it subscribes to, and returns msg_id (the webhook-id header of that delivery). A disabled endpoint answers endpoint_disabled. Free.

Use flatmark from Claude Code, Cursor or Codex

Connect flatmark in LiveContext, create an API key in Settings, then add the LiveContext MCP server to your client. Its flatmark operations become tools the client can call, next to every other integration you connected.

Claude Code

claude mcp add --transport http livecontext https://livecontext.ai/mcp --header "X-API-Key: YOUR_API_KEY"

Codex (~/.codex/config.toml)

[mcp_servers.livecontext]
url = "https://livecontext.ai/mcp"

[mcp_servers.livecontext.http_headers]
"X-API-Key" = "YOUR_API_KEY"

Cursor, Claude Desktop and other JSON configs

{
  "mcpServers": {
    "livecontext": {
      "type": "http",
      "url": "https://livecontext.ai/mcp",
      "headers": {
        "X-API-Key": "YOUR_API_KEY"
      }
    }
  }
}

Set up the MCP server step by step

How to connect flatmark

Create your workspace
Sign up free. Nothing to install.
Connect flatmark once
Add your flatmark connection in Settings. Every tool from this integration reuses it, and secrets are encrypted at rest and visible only to you.
Describe the job in chat
The workflow builds itself in front of you and picks the endpoints it needs. Drag the steps yourself on the canvas whenever you would rather be explicit.