MCP vs. REST API

The MCP server and the REST API overlap heavily, but neither is a strict superset of the other. See Getting started for an API overview.

Surface	MCP	REST API
Agents, versions, skills, artifacts, model groups	Yes	Yes
Agent sessions (start, poll, follow up, cancel)	Yes	Yes
Connected MCP servers (list, inspect, call tools)	Yes	Yes
Flows, workbooks, and runs	Yes	Yes
Company Brain search	Yes	Yes
Personal wiki queries	No	Yes
Audit logs	Yes (`get_audit_logs`, time range only)	Yes (`GET /get_audit_logs`, with event/user/IP/workspace filters)
Organization data exports	Yes (`export_data`, `get_export_status`)	Yes
Gumloop documentation search	Yes (`search_documentation`, `ask_gummie`)	No
Evaluations	Yes	Yes
Session approvals	Yes (`resolve_session_approvals`)	Yes
Queued messages	No	Yes
Session access controls	No	Yes
Session file upload	No	Yes
Workspace file upload and download	No	Yes
Team and custom role membership, custom role credit limits	Yes	Yes
Agent triggers, knowledge sources, and subagents	Yes	Yes
Browser profiles and cookie import	No	Yes
Agent incognito enforcement	No	Yes
Model groups and model routing	Yes	Yes
Chat completions (OpenAI-compatible)	No	Yes

Use MCP when an AI client should manage Gumloop agents, sessions, skills, flows, Brain, evaluations, and organization administration. The REST API also exposes surfaces not available through MCP, including queued messages, session access controls, session and workspace file operations, browser profiles, personal wiki queries, agent incognito enforcement, and OpenAI-compatible chat completions.

Core concepts

Guides

Tools

MCP vs. REST API