calls_make
Place an outbound AUDIO/VOICE phone call via Twilio (PSTN) or Telegram (MTProto 1:1 call). Use this any time the user asks to 'call', 'ring', 'phone', 'dial', or have a spoken conversation. Do NOT use messages.send when the user asks to call someone — a call is real-time voice, not a text message...
This record as markdown: /tools/io-github-saloprj-dialogbrain/calls-make.md
What calls_make does on Dialogbrain
AI agents invoke calls_make to trigger actions in Dialogbrain. What it does depends on the arguments the agent supplies, and its effects often reach beyond the immediate call: builds kicked off, notifications sent, workflows started.
| Parameter | Type | Required | Description |
|---|---|---|---|
app | string | — | Android package that places the call (generic — any app with a dispatch recipe, or manual UI dispatch). Example: 'com.whatsapp'. Used only when channel='android |
channel | string | — | Voice transport: 'twilio' (phone via PSTN — requires phone_number in E.164), 'telegram' (MTProto 1:1 call — requires telegram_user_id, NOT a phone number or thr |
greeting | string | Yes | The first sentence the agent speaks immediately when the call connects. ALWAYS provide a greeting — without it the caller hears silence. Keep it short and natur |
contact_id | integer | — | LINE only: the contact entity ID for storing the confirmed display name after placement. Ignored for other channels/apps. |
report_back | string | — | When to re-invoke you after the call ends. 'on_answer' (default) = only if the call was answered, 'always' = even on missed/failed calls, 'never' = fire and for |
instructions | string | — | What to do during the call — objective, questions, tone. The AI generates a natural opening and guides the conversation. Example: 'Call about invoice #1234. Ask |
phone_number | string | — | Destination phone number in E.164 format (e.g., '+15551234567', '+66812345678'). Required when channel='twilio'. |
maxru_user_id | string | — | Destination Max.ru user/chat ID. Required when channel='maxru'. |
skip_dispatch | boolean | — | Set true after you placed the call manually with the android_* UI tools — skips the automatic recipe and attaches straight to the already-live call. OMIT for th |
voice_agent_id | integer | — | ID of the agent that conducts the call (an `id` from agents.list). If omitted, uses the workspace's default voice-capable agent when one exists. Pass this when |
telegram_user_id | string | — | Destination Telegram user ID (decimal int64 as string, e.g. '123456789'). Required when channel='telegram'. The caller account must have had prior interaction w |
target_display_name | string | — | LINE only: the contact's LINE display name to place the voice call to (resolved by the planner from the contact). Ignored for other channels/apps. |
Parameters from the server's own tool schema.
Why calls_make is rated High
This tool initiates a real-time outbound phone call to an external party, which is an external operation with significant real-world effects. The AI agent conducts the conversation autonomously using provided instructions, meaning misuse could result in unauthorized calls being placed to real people.
From the tool's definition Place an outbound AUDIO/VOICE phone call via Twilio (PSTN) or Telegram (MTProto 1:1 call)... You conduct the conversation as the voice agent using the provided greeting and instructions.
Risk signalsHigh parameter count (12 properties)
Attacks that exploit this kind of access
The rule that runs calls_make safely
PolicyLayer is an MCP gateway: it sits between your AI agents and Dialogbrain, and checks every tool call against a rule you set before the call runs. Nothing changes on the server itself. For calls_make, this is the rule to start with:
calls_make stays usable, but rate-capped: a runaway agent can't fire it dozens of times a minute. Everything else on the server is denied unless you say otherwise.
The button opens the PolicyLayer dashboard: create your workspace, connect Dialogbrain, apply this rule, and every calls_make call is checked against it from then on.
Questions about calls_make
Place an outbound AUDIO/VOICE phone call via Twilio (PSTN) or Telegram (MTProto 1:1 call). Use this any time the user asks to 'call', 'ring', 'phone', 'dial', or have a spoken conversation. Do NOT use messages.send when the user asks to call someone — a call is real-time voice, not a text message. You conduct the conversation as the voice agent using the provided greeting and instructions. It is categorised as a Execute tool in the Dialogbrain MCP Server, which means it can trigger actions or run processes. Use rate limits and argument validation.
calls_make accepts 12 parameters: app, channel, greeting, contact_id, report_back, instructions, phone_number, maxru_user_id, skip_dispatch, voice_agent_id, telegram_user_id, target_display_name. Required: greeting. The full parameter table on this page comes from the server's own tool schema.
Register the Dialogbrain MCP server in PolicyLayer and add a rule for calls_make: allow, deny, rate-limit, or require approval. Point your MCP client at the PolicyLayer proxy URL and the rule is enforced on every call, before it reaches Dialogbrain. Nothing to install.
calls_make is a Execute tool with high risk. Execute tools should be rate-limited and have argument validation enabled.
Yes. Add a rate_limit block to the calls_make rule in your PolicyLayer policy. For example, setting max: 10 and window: 60 limits the tool to 10 calls per minute. Rate limits are tracked per agent session and reset automatically.
Set action: deny in the PolicyLayer policy for calls_make. The AI agent will receive a policy violation error and cannot call the tool. You can also include a reason field to explain why the tool is blocked.
calls_make is provided by the Dialogbrain MCP server (https://api.dialogbrain.com/mcp). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.
More on Dialogbrain, and thousands of servers like it.
Across the catalogue