Skip to main content
Glama
ghl-mcp-server-v2

ghl-mcp-server-v2

by zackscriven

ghl_kb_crawler_train

Queues selected discovered website pages for AI ingestion into the knowledge base after discovery has found URLs.

Instructions

Train discovered website pages and ingest into the knowledge base Queues selected discovered pages so the AI ingests their content into the knowledge base. WHEN TO USE: use this when discoverWebsite has already found pages; use discoverWebsite to crawl first; use deleteTrainedUrlsForKnowledgeBase to untrain pages. RETURNS: whether training was queued, a status message, and the URL ids that were accepted. Endpoint: POST /knowledge-bases/crawler/train (Version header: v3; source: v3/knowledge-base-v3.json) Requires a Sub-Account (location) token. OAuth scopes: knowledge-bases.write

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": {
      "body": "<body>"
    },
    "name": "ghl_kb_crawler_train"
  }
}

Input Schema

NameRequiredDescriptionDefault
bodyYesRequest body (schema carried verbatim from the official OpenAPI spec).

Schema Changelog

Changes observed during successful MCP inspections.

  1. Changed9 schema fields changedv2.1.0
    • changedInput schema / properties / body / properties / knowledgeBaseId / description
      Previous value: -"knowledge base id"New value: +"Knowledge base ID as string"
    • changedInput schema / properties / body / properties / knowledgeBaseId / example
      Previous value: -"jjkkxftgvbhjmn,"New value: +"RXzAdYGknD2phAoln3Jl"
    • changedInput schema / properties / body / properties / locationId / example
      Previous value: -"tDtDnQdgm2LXpyiqYvZ6"New value: +"jNtgzTfHnLKErAWswvVE"
    • changedInput schema / properties / body / properties / operationId / description
      Previous value: -"operation id as string"New value: +"Operation ID as string"
    • changedInput schema / properties / body / properties / operationId / example
      Previous value: -"688b640bcb02d498102a13f0,"New value: +"689267ba9d8d63ea160ee9c7"
    • changedInput schema / properties / body / properties / urlIds / description
      Previous value: -"List of Object ids of the discovered urls"New value: +"List of URL IDs to train"
    • changedInput schema / properties / body / properties / urlIds / example
      Previous value: -[
      -  "688b640bcb02d498102a13ec",
      -  "688b640bcb02d498102a13ea"
      -]New value: +[
      +  "689267ccb27254d92da17b69"
      +]
    • changedInput schema / properties / body / required
      Previous value: -[
      -  "locationId",
      -  "urlIds",
      -  "knowledgeBaseId",
      -  "operationId"
      -]New value: +[
      +  "locationId",
      +  "knowledgeBaseId",
      +  "operationId",
      +  "urlIds"
      +]
    • addedInput schema / properties / body / title
      Added value: +"TrainSelectedUrlsPublicDTO"
  2. First observedv2.0.0

TDQS

A4.2/5.0
Behavior4/5

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

Annotations cover the safety profile (destructiveHint=false, idempotentHint=false, openWorldHint=true), and the description adds genuinely new context: the operation is queued/asynchronous, it requires a Sub-Account (location) token, and it needs the knowledge-bases.write scope. It stops short of saying what happens when URLs are re-trained or how long ingestion takes.

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?

Well front-loaded with labeled WHEN TO USE and RETURNS sections plus endpoint metadata. The opening line largely restates the second sentence, which is a small redundancy but not costly.

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 nested-body mutation tool with no output schema, the description supplies the prerequisites, the auth requirement, the scope, and the shape of the response (queued flag, status message, accepted URL ids), which is everything an agent needs to invoke it correctly.

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 100%, with per-field descriptions and examples for locationId, knowledgeBaseId, operationId, and urlIds. The description only implies that urlIds come from a prior discovery run and that accepted ids are echoed back; it adds little beyond the schema, so the baseline of 3 applies.

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?

States a specific verb and resource — queueing discovered website pages for ingestion into a knowledge base — and separates it from the discovery and untrain siblings by name. The second sentence also clarifies that the tool only queues work rather than performing the crawl itself.

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?

The WHEN TO USE block gives a clear prerequisite (discovery must have run first) and names the untrain alternative. However, the referenced names ('discoverWebsite', 'deleteTrainedUrlsForKnowledgeBase') do not match the actual sibling tool names (ghl_kb_crawler_discover, ghl_kb_crawler_delete_urls), so the agent must infer the mapping.

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