Skip to main content
Glama

manage_recruitment

Destructive

Act on the survey's live study at Prolific or CloudResearch Connect, using the API key saved in the app. One action per call: create the draft study or unlaunched project, change what participants see, pause, stop, close, or discard it, reject or report participants, block them from future Connect projects, message them, save participant groups, or unlink. Changes happen at the provider and some can't be undone. Nothing here pays participants: launching, approving, bonuses, and automatic approval are app-only. The prolific_* actions (and read_deploy's prolific_* includes) are in beta and need Beta Tester access; connecting Prolific by study link and completion code in edit_deploy doesn't. Read state back with read_deploy.

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": {
      "survey_id": "<survey_id>"
    },
    "name": "manage_recruitment"
  }
}

Input Schema

NameRequiredDescriptionDefault
survey_idYesSurvey UUID (its recruitment integration must already be connected with edit_deploy recruitment.connect)
prolific_reviewNoProlific only: reject submissions or ask for returns — the non-payment actions of the Prolific column in the Export response table.
prolific_unlinkNoProlific only: forget the linked Prolific study (the study on Prolific is untouched).
prolific_updateNoProlific only: pause or stop the linked study, or cancel its scheduled publish. Read it back with read_deploy include ["prolific_study"].
prolific_messageNoProlific only: message participants of the linked study through Prolific.
prolific_save_groupNoProlific only: save participants as a Prolific participant group, to invite (or leave out) in a later study.
cloudresearch_reviewNoCloudResearch Connect only: reject or flag assignments — the non-payment actions of the CloudResearch column in the Export response table.
cloudresearch_unlinkNoCloudResearch Connect only: forget the linked Connect project (the project on Connect is untouched).
cloudresearch_updateNoCloudResearch Connect only: pause, close, or discard the linked Connect project, or cancel its scheduled launch. Read it back with read_deploy include ["cloudresearch_project"].
prolific_set_detailsNoProlific only: change what participants see on the linked study. Prolific lets a draft change all three; once published, only the description. Read it back with read_deploy include ["prolific_study"].
cloudresearch_messageNoCloudResearch Connect only: message participants of the linked project through Connect.
prolific_create_studyNoProlific only: create the study on Prolific as an unpublished draft with the survey's study link, completion codes (made here when the survey has none, and written into the Complete end screen's redirect), audience, quotas, and waves; it always leaves out the "Domandata block list" group (see prolific_block_participants), and returns never_twice saying who else was left out. Nothing is charged until it's published; publishing spends the Prolific balance, so the researcher publishes it in the app (Deploy → Recruitment → Prolific) once the survey is published with that change.
cloudresearch_save_groupNoCloudResearch Connect only: save participants as a Connect participant group, to invite (or leave out) in a later project.
cloudresearch_set_detailsNoCloudResearch Connect only: change what participants see on the linked project — title, description, length, auto-pay window, time limit, instructions, requirements, sensitive-content flag, and language. Connect allows this only before launch. Read it back with read_deploy include ["cloudresearch_project"].
prolific_block_participantsNoProlific only: add the participants behind these responses to the account's "Domandata block list" participant group (made if there is none). Every new study created from Domandata leaves that group out (prolific_targeting.block_list shows its size); studies made on Prolific itself don't, unless the group is added there as a blocklist screener. Remove people from the group on Prolific to undo.
cloudresearch_create_projectNoCloudResearch Connect only: create the Connect project (unlaunched — nothing is charged) with the survey's study link, the payment, audience, and number of responses, then point the Complete end screen at the project's own return link. Launching spends the Connect balance, so the researcher launches it in the app (Deploy → Recruitment → CloudResearch Connect) once the survey is published with that change.
cloudresearch_block_participantsNoCloudResearch Connect only: add the participants behind these responses to the account's universal block list ("Domandata block list", made if there is none), which keeps them out of every future project, including ones made on Connect. Remove people on Connect to undo.

Output Schema

NameRequiredDescriptionDefault
resultYes

Schema Changelog

Changes observed during successful MCP inspections.

  1. First observed

TDQS

A4.8/5.0
Behavior5/5

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

Annotations already declare destructive/open-world/non-idempotent, but the description adds materially: changes occur at the provider and 'some can't be undone', no payment operations exist here, and prolific_* actions require Beta Tester access. It also constrains dispatch to 'one action per call', which the flat schema of 17 optional objects does not itself make obvious.

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

Conciseness4/5

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

Front-loaded with purpose, then the action list, then behavioral caveats, then routing — a sensible order with no filler sentences. The middle action enumeration is a long run-on list, though it is largely necessary for a multi-action dispatch tool.

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?

For a 17-parameter polymorphic tool with an output schema present, the description covers the essentials: provider scope, one action per call, irreversibility, the app-only payment operations, the beta gate, and where to read state back. Return values need not be explained given the output schema.

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

Parameters4/5

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

Schema description coverage is 100%, so the schema carries nearly all parameter meaning; the description's main added value is the 'one action per call' rule and the note that the API key is the one saved in the app. It does not restate per-action semantics, which is appropriate, but it could have clarified what happens if multiple action objects are supplied.

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?

The opening sentence names the resource ('the survey's live study at Prolific or CloudResearch Connect') and the verb family ('Act on'), then enumerates every concrete action (create draft, change details, pause/stop/close/discard, reject/report, block, message, save group, unlink). It explicitly differentiates from siblings by naming read_deploy for state and edit_deploy for connecting a study. An agent can tell this is the write-side recruitment tool without opening the 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?

It routes clearly: use read_deploy to read state back, use edit_deploy to connect Prolific by study link/completion code, and use the app (not this tool) for launching, approving, bonuses, and automatic approval. It also flags the beta gate on prolific_* actions versus the non-beta edit_deploy path, which is exactly the kind of when-to-use condition an agent needs.

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

Try in Browser

Glama MCP Gateway

Add one secure layer between your agents and this server.

Resources