Skip to main content
Glama
KaiCalls MCP Server

KaiCalls MCP Server

Official

Get Coaching

get_coaching
Read-onlyIdempotent

Retrieves coaching feedback for AI phone agents: per-call strengths and fixes with exact quotes and timestamps, or recurring habits across recent calls.

Instructions

Coach the business's agent. With call_id: what the agent did well and what to change on that call, each point with the caller's or agent's exact words, the second it happened, and a one-line rule that would fix it (teach it with teach_coaching_rule). Without call_id: the agent's recurring habits across recent calls (how often, example calls, and whether a rule taught for it has worked). Read-only.

Tool call template

This tool declares no required parameters. Add optional parameters from the input schema as needed.

{
  "id": 1,
  "jsonrpc": "2.0",
  "method": "tools/call",
  "params": {
    "arguments": {},
    "name": "get_coaching"
  }
}

Input Schema

NameRequiredDescriptionDefault
daysNoWindow for habits, in days (default 30).
call_idNoCoach one call. Omit for habits across calls.
agent_idNoHabits for one agent only (optional).
business_idNoBusiness ID. Required when the key can reach more than one business.

Output Schema

NameRequiredDescriptionDefault
errorNoPresent when success is false
habitsNo
successYesWhether the tool completed successfully
coachingNo

Schema Changelog

Changes observed during successful MCP inspections.

  1. First observedv0.1.0

TDQS

A4.2/5.0
Behavior3/5

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

Annotations already declare readOnlyHint, idempotentHint, destructiveHint=false and closed-world, and the trailing "Read-only." merely restates them. The remaining description is almost entirely a preview of return contents (what went well, exact quotes, timestamps, whether a rule worked), which the output schema already covers, so little genuinely new behavioral context is added beyond the business_id reach note.

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

Conciseness5/5

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

Two sentences, heavily front-loaded with the conditional structure ("With call_id… Without call_id…") so the most decision-relevant information comes first. Parenthetical asides still carry actionable content rather than padding.

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?

With an output schema present, the description needn't enumerate return fields, and it still covers both modes, the parameters' roles, and the follow-up action. The only soft spot is the absence of any sense of result volume or limits for the habits mode, which is minor for a read-only tool.

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 baseline is 3, but the description earns above baseline by explaining the pivotal semantic: call_id is a mode switch, not just a filter, and days/agent_id only shape the habits mode. It also flags business_id as conditionally required when the key spans multiple businesses, which clarifies the schema's vaguer phrasing.

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 description opens with a specific verb and resource ("Coach the business's agent") and then splits the tool's two distinct behaviors by the presence of call_id, so an agent can immediately tell which mode it will get. This is far more specific than the sibling get_* tools, which return raw data rather than coaching output.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Usage Guidelines4/5

Does the description explain when to use this tool, when not to, or what alternatives exist?

It gives explicit routing: pass call_id to coach a single call, omit it to see habits across recent calls, and use teach_coaching_rule to act on a finding. What's missing is any exclusion guidance — nothing says when to prefer get_analytics or list_recent_calls over this tool — but the internal mode selection is unambiguous.

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

Deploy Server

Other Tools