generated: '2026-07-18'
method: derived
status: candidate
source: https://docs.corti.ai/llms.txt
description: Corti does not publish a hosted/remote MCP server for its own API; instead the Corti Agentic
  Framework acts as an MCP *client*, registering and calling external MCP servers (streamable_http transport,
  authorizationType none|inherit) as "experts". The tool list below is a CANDIDATE derived from Corti's
  documented API operations to illustrate an MCP surface an integrator could expose.
server:
  name: corti
  transport: http
  note: candidate only — not an official Corti-hosted MCP endpoint
related:
  agentic_mcp_authentication: https://docs.corti.ai/agentic/mcp-authentication
  a2a_protocol: https://docs.corti.ai/agentic/a2a-protocol
tools:
- name: predict_codes
  description: Predict ICD-10/CPT/OPCS-4/etc. medical codes from clinical text
  source: https://docs.corti.ai/api-reference/codes/predict-codes
- name: generate_document
  description: Generate a clinical document from an interaction
  source: https://docs.corti.ai/api-reference/documents-classic/generate-document
- name: generate_structured_document
  description: Generate a structured (guided) document from a template
  source: https://docs.corti.ai/api-reference/guided-documents/generate-a-structured-document
- name: extract_facts
  description: Extract facts from provided text without storing them
  source: https://docs.corti.ai/api-reference/facts/extract-facts
- name: list_facts
  description: Retrieve facts for a given interaction
  source: https://docs.corti.ai/api-reference/facts/list-facts
- name: send_message_to_agent
  description: Send a message to a Corti agent to start or continue a task
  source: https://docs.corti.ai/agentic/agents/send-message-to-agent
- name: list_agents
  description: List agents callable by the Corti Agent Framework
  source: https://docs.corti.ai/agentic/agents/list-agents
deployment:
  mode: none
  verified: derived
  tools: 7
  checked: '2026-08-12'
  source: catalog MCP census