Frank

Create a research in Frank

create_research

Creates a research study in Frank as a draft. It is not live and no participant can reach it until publish_research is called, which the user approves separately. Confirm the goals and topics with the user before calling this — never assemble them from an interview transcript.

Tool call template

Replace each placeholder with a value matching the input schema. Only required parameters are included; add optional parameters as needed.

{
  "id": 1,
  "jsonrpc": "2.0",
  "method": "tools/call",
  "params": {
    "arguments": {
      "name": "<name>",
      "description": "<description>",
      "goals": "<goals>",
      "topics": "<topics>"
    },
    "name": "create_research"
  }
}

Input Schema

Name	Required	Description
`name`	Yes	What this study is called.
`goals`	Yes	What the user wants to learn. At least one.
`topics`	Yes	The interview structure: sections, each with its questions.
`prompts`	No
`botTypes`	No	How interviews are conducted: audio, video or chat.
`description`	Yes	What the study is about.
`researchType`	No
`greetingMessage`	No

Schema Changelog

Changes observed during successful MCP inspections.

First observedOct 30, 2026

TDQS

A4.4/5.0

Behavior4/5

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

Annotations already declare this is a non-read-only, non-destructive, non-idempotent write. The description adds genuinely new behavioral context: the created object is a draft, is unreachable by participants, and only becomes live through a separately approved publish call. It stops short of saying what happens on repeated invocations (duplicate studies) despite idempotentHint=false, which keeps it from 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 short sentences, zero filler, and the critical scoping fact (draft, not live) is front-loaded before the procedural caveat. Every sentence earns its place.

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

Completeness4/5

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

For a mutation tool with a deeply nested schema, no output schema, and 8 params, the description covers lifecycle, approval flow, and the most error-prone inputs. It leaves the agent without guidance on the optional configuration params (botTypes, researchType, prompts), which is the remaining gap.

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?

At 63% schema coverage across 8 params, the schema carries most of the load. The description adds real meaning for goals and topics by requiring user confirmation rather than transcript-derived assembly, but prompts, botTypes, researchType, and greetingMessage are unexplained in both places. That is adequate, not compensating.

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 ("Creates a research study in Frank") plus the lifecycle state ("as a draft"), which immediately distinguishes it from publish_research and edit_research. An agent can route correctly without opening any schema.

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 names the downstream alternative (publish_research, with separate user approval) and states the precondition for calling this tool: confirm goals and topics with the user first. It also gives a hard exclusion ("never assemble them from an interview transcript"), which is exactly the kind of when-not guidance that prevents misfires.

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

All Frank tools Inspect

Try in Browser

One endpoint for every MCP toolRayrun runs every MCP server behind one URL and holds the credentials. Each call is decided per client, tool, and argument, then recorded.ray.runAd

Glama MCP Gateway

Add one secure layer between your agents and this server.

Resources

Need Help?