LlamaCloud MCP server

What the model can do

LlamaCloud (LlamaParse) document processing for a project: parse, extract and classify jobs with status, tier, errors, credits spent and a classification's document type, managed indexes and pipelines with their sync state, files by type and expiry, and credit usage by day, event type or tier; when you ask, can parse a document from a file or public URL, extract structured fields or classify it (each spends credits), cancel or delete a job, sync an index or cancel its sync, and delete an index or a file.

Reads

• llamacloud_list_parse_jobsLlamaParse parse jobs in the project: id, status, tier, error and credits, with counts by status and credits over what was read.
• llamacloud_list_extract_jobsLlamaExtract jobs in the project, newest first: id, status, the file or parse job it read, configuration, error and credits.
• llamacloud_list_classify_jobsClassify jobs in the project: id, status, the file it read, the document type it was classified as with confidence, error and credits.
• llamacloud_get_jobOne parse, extract or classify job's current state: status, tier, error, credits, and for classify the document type and confidence.
• llamacloud_list_indexesManaged indexes in the project (documents parsed and chunked for retrieval): name, description, when last synced and exported…
• llamacloud_list_pipelinesRetrieval pipelines in the project, newest first: name, type (managed or playground) and status.
• llamacloud_list_filesFiles uploaded to the project: id, type, purpose, last modified and when it expires.
• llamacloud_usageCredits the organization spent, grouped by day, event type (parse, extract and so on), tier or project, between two days (default the last 30).

Acts, when you ask

These change something in LlamaCloud. The model is told to use them only when you ask for exactly that, and each one is granted by name: per team on a shared connection, by the owner on a personal one.

• llamacloud_parse_documentStarts a LlamaParse job on an uploaded file (file_id) or a public document URL (source_url), at the tier chosen: fast (rule-based, cheapest)…
• llamacloud_extract_documentStarts a LlamaExtract job pulling structured fields out of a file (dfl-...) or a completed parse job (pjb-...)…
• llamacloud_classify_documentStarts a classify job sorting a file (dfl-...) or parse job (pjb-...) into one of the document types given as rules…
• llamacloud_cancel_jobCancels a running LlamaCloud parse, extract or classify job; it is marked CANCELLED and cannot be resumed.
• llamacloud_delete_jobPermanently deletes a finished (completed, failed or cancelled) LlamaCloud parse, extract or classify job and its results.
• llamacloud_sync_indexSyncs a LlamaCloud index now: changed source files are re-parsed and updated chunks exported, which spends credits.
• llamacloud_cancel_index_syncCancels a LlamaCloud index's running sync.
• llamacloud_delete_indexPermanently deletes a LlamaCloud index; anything retrieving from it stops finding its documents.
• llamacloud_delete_filePermanently deletes a file from the LlamaCloud project.

Connect it

First connect LlamaCloud on Dashboard → Connections, then add Memtro to your client once. The LlamaCloud tools appear alongside the tools of everything else you have connected; nothing is configured per connector.

claude plugin marketplace add memtro-org/claude-memtro
claude plugin install memtro@memtro --config api_key=sk-memtro-...

{
  "mcpServers": {
    "memtro": {
      "url": "https://platform.memtro.com/api/mcp",
      "headers": {
        "Authorization": "Bearer sk-memtro-..."
      }
    }
  }
}

{
  "mcp": {
    "memtro": {
      "type": "remote",
      "url": "https://platform.memtro.com/api/mcp",
      "headers": {
        "Authorization": "Bearer sk-memtro-..."
      }
    }
  }
}

Who may use it

The key identifies the person, and the server offers that person the tools they may use everywhere else in Memtro. A shared LlamaCloud connection is used by the teams an administrator granted it to, tool by tool, on the Permissions page; a personal connection is used by its owner alone, and its tools that act stay off until the owner allows them. See Teams and permissions and Connectors.

Ask it like this

• › How many credits did parsing use this month?
• › Which parse jobs failed this week, and why?
• › When was the contracts index last synced?
• › What document types did yesterday's classify jobs find?
• › Sync the contracts index now.

Questions

Is this an official LlamaCloud MCP server?

No. It is Memtro's MCP server with Memtro's LlamaCloud connector behind it: the connector reads LlamaCloud through its API with the credentials you connect on Dashboard → Connections, and LlamaCloud publishes nothing here. The same endpoint carries every other system you connect and Memtro's memory tools.

What does it cost?

Nothing on top of Memtro. One price per organisation by its number of users, from $50 a month for up to 10 users after a 7-day free trial, with every connector, the MCP server, jobs and memory included. You bring your own model keys.

Can it change anything in LlamaCloud?

Only when you ask, and only what it has been granted: parse a document from an uploaded file or public URL (spends credits per page); extract structured fields from a document (spends credits per page); classify a document against rules (spends credits); cancel a running parse, extract or classify job; delete a finished job and its results; sync an index, re-parsing changed files (spends credits); cancel an index's running sync; delete an index; delete a file. Every other tool reads, searches or lists. An administrator grants each acting tool by name per team, and a personal connection acts only after its owner allows it.

Which clients work?

Any MCP client that speaks Streamable HTTP: Claude Code, Cursor, Claude Desktop, OpenCode, Codex, Zed, Windsurf, Cline and the Anthropic and OpenAI SDKs. There is no local process to run; the server is remote and authenticated with your key.

Other MCP servers · Engineering and operations

• GitHub MCP server
• AWS CloudWatch MCP server
• Sentry MCP server
• ServiceNow MCP server
• Okta MCP server
• Twilio MCP server