Skip to main content
Glama

Start a broadcast

dosai_start_broadcast
Destructive

Send one message to many clients of this project at once. REAL PEOPLE GET IT AND IT CANNOT BE UNSENT: show the owner the exact text and the audience numbers from preview_broadcast_audience, and start only after they say yes. Only people who have written to the project at least once are included; the platform paces the sending (about one message per 30-45 seconds per number), stops at the daily cap, respects working hours and skips anyone who unsubscribed. For WhatsApp the owner must confirm they have consent (consent_confirmed). Give 2-4 variants of the text, or turn on paraphrase, so the same wording does not go out in a fan: identical mass texts are what gets a number banned.

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": {
      "project_id": "<project_id>",
      "body": "<body>"
    },
    "name": "dosai_start_broadcast"
  }
}

Input Schema

NameRequiredDescriptionDefault
tagNo
bodyYesThe message, exactly as people will see it. {{name}} is replaced with the client's name.
titleNoName of the broadcast, for the owner's report only.
channelNoDefault: telegram.
media_urlNoPicture to send with the text (a link from the project media library).
paraphraseNoLet the AI reword every message (costs tokens, about 11 cents per 220 messages).
project_idYesFrom dosai_list_projects.
lead_statusNo
window_daysNo
scheduled_atNoISO time to start later. Omit to start now.
body_variantsNoUp to 4 more wordings of the SAME message. They rotate between recipients.
wa_template_idNoOfficial WhatsApp: approved MARKETING template for people silent over 24 hours; Meta charges the owner for each.
consent_confirmedNoRequired for WhatsApp: the owner confirms these people agreed to receive messages.

Schema Changelog

Changes observed during successful MCP inspections.

  1. Changed1 schema field changed
    • addedInput schema / properties / wa_template_id
      Added value: +{
      +  "description": "Official WhatsApp: approved MARKETING template for people silent over 24 hours; Meta charges the owner for each.",
      +  "format": "uuid",
      +  "type": "string"
      +}
  2. Changed2 schema fields changed
    • removedInput schema / $schema
      Removed value: -"http://json-schema.org/draft-07/schema#"
    • changedInput schema / properties / project_id / description
      Previous value: -"Project id, as returned by dosai_list_projects."New value: +"From dosai_list_projects."
  3. Changed1 schema field changed
    • removedInput schema / properties / project_id / pattern
      Removed value: -"^([0-9a-fA-F]{8}-[0-9a-fA-F]{4}-[1-8][0-9a-fA-F]{3}-[89abAB][0-9a-fA-F]{3}-[0-9a-fA-F]{12}|00000000-0000-0000-0000-000000000000|ffffffff-ffff-ffff-ffff-ffffffffffff)$"
  4. Added

TDQS

A4.4/5.0
Behavior4/5

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

Well beyond the annotations, it discloses that messages reach real people and cannot be unsent, that sending is paced at roughly one message per 30-45 seconds per number, that it halts at the daily cap, respects working hours, and skips unsubscribed contacts. It also notes Meta charges per template message. What it omits is what the call returns (a broadcast id for later resume/stop) and how to recover once started.

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?

The irreversible-consent warning is correctly front-loaded in capital letters, and the rest is dense with operational fact rather than filler. It is on the long side for a tool description, but nearly every clause carries a distinct constraint the agent would otherwise get wrong.

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 13-parameter destructive, open-world tool with no output schema, the description covers the risk profile, prerequisites, audience eligibility, rate-limit behavior, WhatsApp consent, and cost implications. The remaining gaps (return value, tag/lead_status/window_days semantics) are minor compared to the hazards it addresses.

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 coverage is 77%, so the baseline is 3, and the description largely restates schema content: the consent_confirmed meaning and the 11-cent paraphrase cost are already documented in the schema. It adds a little connective meaning (variants and paraphrase are alternatives for de-duplicating wording) but leaves tag, lead_status, and window_days unexplained in both places.

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?

"Send one message to many clients of this project at once" gives a specific verb, resource, and blast radius, and the following sentences pin down the audience filter (only people who wrote at least once). It is clearly distinguishable from dosai_preview_broadcast_audience, which it explicitly routes the agent to first.

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 states an explicit precondition chain: run preview_broadcast_audience, show the owner the exact text and audience numbers, and start only after they say yes. It also names the WhatsApp-specific gate (consent_confirmed) and the anti-ban requirement of 2-4 variants or paraphrase, so the agent knows the conditions rather than inferring them.

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