analyze_voice_instruction_gaps is a tool in the Speak AI MCP server. Advisory
only: compares a voice agent’s current instructions against anchors/original
intent/recent call summaries you supply and suggests up to 3 patches.
What does it do?
Advisory only: compares a voice agent’s current instructions against
anchors/original intent/recent call summaries you supply and suggests up to 3
patches. Requires the OWNER or ADMIN role. Nothing is written; pass a suggestion’s
suggestedPatch to apply_voice_instruction_gap to actually apply it.
Parameters
analyze_voice_instruction_gaps takes 4 parameters, 1 of them required.
| Name | Type | Required | Description |
|---|---|---|---|
agentId |
string | Yes | ID of the voice agent (from list_voice_agents) |
anchors |
array of string | No | Specific requirements the instructions must cover. Defaults to empty. |
originalPrompt |
string | No | The original generation prompt, for context. |
conversationSummaries |
array of string | No | Recent call summaries, to ground suggestions in what actually came up. Defaults to empty. |
Behaviour
The Speak AI MCP server publishes these behaviour hints for
analyze_voice_instruction_gaps.
- Only reads data. Does not modify your workspace.
- Does not delete or overwrite existing data.
- Repeating an identical call can have an extra effect.
- Stays inside Speak AI. Does not reach other systems.
Related
- Voice agent insights lists the rest of the Voice agent insights category.
- Tool reference lists every Speak AI MCP server tool.
- MCP server covers setup, supported clients, and what agents can do once connected.