deer-flow-mcp

Start Deep Research

deerflow_research

Kick off a long-running deep research run on a topic and get thread/run IDs plus a polling URL; fetch cited findings on completion.

Instructions

Kick off a long-running deep-research run on a fresh DeerFlow thread. Returns immediately with the thread/run ids and a web URL; poll deerflow_run_status until it reaches a terminal status, then call deerflow_get_report to read the findings. Deep research takes minutes to ~45 minutes and never blocks this call.

Input Schema

TableJSON Schema

Name	Required	Description
`focus`	No	Optional one-line constraint to fold into the brief (e.g. 'focus on the EU').
`model`	No	Optional DeerFlow model name (see deerflow_list_models).
`topic`	Yes	The research topic or question to investigate in depth.
`recursion_limit`	No	Optional agent recursion budget for this run (default 1000).

Output Schema

TableJSON Schema

Name	Required	Description	Default
`run_id`	Yes
`status`	Yes
`web_url`	Yes
`thread_id`	Yes

Schema Changelog

Changes observed during successful MCP inspections.

First observedv0.1.5Sep 12, 2026

TDQS

A4.5/5.0

Behavior4/5

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

Annotations declare this is not read-only, is open-world, and not idempotent, but don't convey timing. The description adds crucial behavior: it returns immediately, never blocks, can take minutes to ~45 minutes, and yields thread/run ids plus a web URL. It doesn't state auth requirements or what happens on failure, keeping it just short of a 5.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness5/5

Is the description appropriately sized, front-loaded, and free of redundancy?

Three sentences, front-loaded with the action, then the async lifecycle, then the timing guarantee. No filler; each sentence earns its place.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Completeness5/5

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

Output schema exists so return values needn't be explained, yet the description still tells the agent what ids/URL come back and which siblings to call next. The full lifecycle is covered for an async launch tool.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Parameters3/5

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

Schema description coverage is 100%, so all four parameters (topic, focus, model, recursion_limit) are already documented inline. The description adds no parameter-level detail beyond the schema, which is the expected baseline when the schema does the work.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Purpose5/5

Does the description clearly state what the tool does and how it differs from similar tools?

States a specific verb and resource ('kick off a long-running deep-research run on a fresh DeerFlow thread'), distinguishing it from deerflow_chat and the polling/report siblings. An agent can immediately tell this is the entry point that starts a run.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Usage Guidelines5/5

Does the description explain when to use this tool, when not to, or what alternatives exist?

Explicitly lays out the workflow chain: this call returns ids, then poll deerflow_run_status until terminal, then call deerflow_get_report. It also names when to expect it to be used (deep research, minutes to ~45 min) versus a blocking call. Alternatives are named directly.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.

Deploy Server