CallMCP — Phone Call MCP Server

Update Agent Configuration

update_agent_config

DestructiveIdempotent

Edit an agent's live runtime configuration — greeting/first message, inbound or SMS prompt, the after-call summary and lead-score instructions, voice, language model, max call duration, and call-transfer settings — routed through the governed update broker so every change keeps the consent + audit trail (a versioned config snapshot and change history). Read get_agent_config first; inbound_prompt replaces the whole prompt. Each applied change is saved as a version (list_config_versions); rollback_config can restore a version's greeting, voice, prompt, model and summary/scoring, but not transfer settings or call length. Mutating changes need an idempotency_key. When the business owner or an admin is signed in through this connector, their own changes apply immediately. High-impact live changes made with an API key or on someone else's behalf queue for the owner's review instead (the owner is notified with a review link). configure_agent_business_rules edits a single named operational-rules section.

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": {
      "agent_id": "<agent_id>"
    },
    "name": "update_agent_config"
  }
}

Input Schema

Name	Required	Description
`name`	No	Agent display name.
`actor`	No	Do not set. Kept for older clients; the signed-in connection identifies you.
`model`	No	Language model selection: { provider, model, temperature }.
`voice`	No	Voice selection: { provider, voiceId }. Use voice_catalog_id instead when you can.
`reason`	No	Why this change is being made. Kept in the change history.
`dry_run`	No	Validate and summarize the change without writing.
`agent_id`	Yes	KaiCalls agent ID to update.
`authority`	No	Do not set. It cannot approve a change: the owner approves through review_url or by being signed in.
`sms_prompt`	No	SMS reply prompt.
`source_ref`	No	External source reference, such as a support ticket or conversation ID.
`business_id`	No	Business ID. Required when the agent has multiple accessible assignments.
`max_duration`	No	Maximum call duration in seconds.
`first_message`	No	Spoken greeting / first message callers hear.
`inbound_prompt`	No	The whole inbound prompt; it replaces the current one. Start from inbound_prompt in get_agent_config, never from a saved version.
`idempotency_key`	No	Required for writes. Reuse the same key when retrying the same change.
`outbound_prompt`	No	Outbound (campaign) system prompt.
`transfer_enabled`	No	Enable/disable live call transfer.
`voice_catalog_id`	No	Preferred way to pick a voice: an id from list_voices.
`queue_for_approval`	No	When authority is missing, create a pending dashboard approval instead of returning needs_approval.
`transfer_phone_number`	No	Number to transfer qualified calls to.
`lead_score_instructions`	No	How each caller is scored from 0 to 100, in plain language. Changes only the scoring. End it with "Respond with ONLY a number from 0 to 100."
`call_summary_instructions`	No	How the summary written after each call should read, in plain language (for example "Write exactly two sentences for the attorney..."). Changes only the summary; the prompt and scoring stay as they are. get_operational_settings shows the current text.

Output Schema

Name	Required	Description
`code`	No
`error`	No	Present when success is false
`result`	No
`status`	No
`message`	No
`success`	Yes	Whether the tool completed successfully
`summary`	No
`agent_id`	No
`approval`	No
`request_id`	No
`risk_level`	No
`business_id`	No

Schema Changelog

Changes observed during successful MCP inspections.

Changed8 schema fields changedNov 12, 2026

changedInput schema / properties / actor / description

Previous value: -"Who is asking: { type: agent|human|system|integration, id, display_name }."New value: +"Do not set. Kept for older clients; the signed-in connection identifies you."

changedInput schema / properties / authority / description

Previous value: -"Attribution only. Caller-supplied authority cannot authorize live changes; authenticated owner review of the stored request is required."New value: +"Do not set. It cannot approve a change: the owner approves through review_url or by being signed in."

addedInput schema / properties / call_summary_instructions

Added value: +{
+  "description": "How the summary written after each call should read, in plain language (for example \"Write exactly two sentences for the attorney...\"). Changes only the summary; the prompt and scoring stay as they are. get_operational_settings shows the current text.",
+  "type": "string"
+}

changedInput schema / properties / inbound_prompt / description

Previous value: -"Full inbound system prompt that drives the agent."New value: +"The whole inbound prompt; it replaces the current one. Start from inbound_prompt in get_agent_config, never from a saved version."

addedInput schema / properties / lead_score_instructions

Added value: +{
+  "description": "How each caller is scored from 0 to 100, in plain language. Changes only the scoring. End it with \"Respond with ONLY a number from 0 to 100.\"",
+  "type": "string"
+}

addedInput schema / properties / reason

Added value: +{
+  "description": "Why this change is being made. Kept in the change history.",
+  "type": "string"
+}

changedInput schema / properties / voice / description

Previous value: -"Voice selection: { provider, voiceId }."New value: +"Voice selection: { provider, voiceId }. Use voice_catalog_id instead when you can."

addedInput schema / properties / voice_catalog_id

Added value: +{
+  "description": "Preferred way to pick a voice: an id from list_voices.",
+  "type": "string"
+}

First observedNov 12, 2026

TDQS

A4.5/5.0

Behavior4/5

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

Goes well beyond the annotations by disclosing the governed update broker, the consent + audit trail, that every write produces a versioned snapshot, that rollback cannot restore transfer settings or call length, and that an idempotency_key is required. The remaining gap is the failure/response surface: it never says what comes back on validation errors or on needs_approval, and dry_run handling is left implicit even though an output schema exists.

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 and then layered into workflow, versioning, and approval routing, with no filler sentences. The opening sentence is a long em-dash-delimited list that is harder to scan than the rest, keeping it short of a 5.

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 22-parameter, destructive, approval-gated mutation, the description covers the critical decision points: pre-read requirement, replacement semantics, versioning, rollback limits, and who can apply versus queue. It stops short of describing the dry_run validation path and partial-failure behavior, which an agent doing a high-impact live edit would want.

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 coverage is 100%, so the baseline is 3; the description still adds real meaning by explaining that inbound_prompt replaces the entire prompt and must be sourced fresh, and by reiterating the mandatory idempotency_key for writes. Several documented parameters (dry_run, queue_for_approval, business_id, source_ref) are not addressed in the prose, so it does not fully supersede the schema.

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?

Opens with a specific verb and resource ("Edit an agent's live runtime configuration") and then enumerates the exact editable surfaces: greeting, inbound/SMS prompt, summary and lead-score instructions, voice, model, max duration, transfer settings. It is distinguishable from siblings such as configure_agent_business_rules, which is scoped to a single named operational-rules section.

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 says to "Read get_agent_config first" and warns that inbound_prompt must be built from that read, never from a saved version. It also names the adjacent tools and their roles (list_config_versions for history, rollback_config for restore, configure_agent_business_rules for rules sections) and explains when a change applies immediately versus queues for owner review.

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

All CallMCP — Phone Call MCP Server 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.