Internal REST API query tool for chat agents providing discovery, monitoring, and troubleshooting capabilities.

Overview

The api_query tool enables chat agents to query the Data Machine REST API (via rest_do_request) for discovery and monitoring. It supports single requests and batch requests.

Parameters

Single request

• endpoint (string, required): REST API endpoint path (e.g., /datamachine/v1/handlers)
• method (string, optional): HTTP method (defaults to GET)
• data (object, optional): Request body data for POST, PUT, or PATCH

Batch requests

• requests (array): Array of requests: { endpoint, method, data?, key? }. If key is omitted, a key is derived from the endpoint path.

Available Endpoints

Discovery

• GET /datamachine/v1/handlers – List all handlers
• GET /datamachine/v1/handlers?step_type={fetch|publish|update} – Filter by type
• GET /datamachine/v1/handlers/{slug} – Handler details and config schema
• GET /datamachine/v1/auth/{handler}/status – Check OAuth connection status
• GET /datamachine/v1/providers – List AI providers and models
• GET /datamachine/v1/tools – List available AI tools

Pipelines (Read-Only)

• GET /datamachine/v1/pipelines – List all pipelines
• GET /datamachine/v1/pipelines/{id} – Get pipeline details with steps and flows

Flows (Read-Only)

• GET /datamachine/v1/flows – List all flows
• GET /datamachine/v1/flows/{id} – Get flow details
• GET /datamachine/v1/flows/problems – List flows flagged for review due to consecutive failures/no items

Jobs & Monitoring

• GET /datamachine/v1/jobs – List all jobs
• GET /datamachine/v1/jobs?flow_id={id} – Jobs for specific flow
• GET /datamachine/v1/jobs?status={pending|processing|completed|failed|completed_no_items|agent_skipped} – Filter by status.
• GET /datamachine/v1/jobs/{id} – Job details

Logs

• GET /datamachine/v1/logs/content – Get log content
• GET /datamachine/v1/logs/content?job_id={id} – Logs for specific job
• DELETE /datamachine/v1/logs – Clear logs
• PUT /datamachine/v1/logs/level – Set log level

System

• GET /datamachine/v1/settings – Get plugin settings
• PATCH /datamachine/v1/settings – Update settings (partial)

Files

• GET /datamachine/v1/files – List uploaded files
• POST /datamachine/v1/files – Upload file
• DELETE /datamachine/v1/files/{filename} – Delete file

Usage Examples

List All Handlers

{
  "endpoint": "/datamachine/v1/handlers",
  "method": "GET"
}

Check OAuth Status

{
  "endpoint": "/datamachine/v1/auth/twitter/status",
  "method": "GET"
}

Get Pipeline Details

{
  "endpoint": "/datamachine/v1/pipelines/123",
  "method": "GET"
}

Monitor Job Status

{
  "endpoint": "/datamachine/v1/jobs/456",
  "method": "GET"
}

Response Format

Single request response

{
  "success": true,
  "data": { /* response body */ },
  "status": 200,
  "tool_name": "api_query"
}

Batch request response

{
  "success": true,
  "batch": true,
  "data": {
    "handlers": { /* ... */ },
    "pipelines": { /* ... */ }
  },
  "errors": {
    "jobs": "Missing endpoint"
  },
  "partial": true,
  "request_count": 3,
  "success_count": 2,
  "error_count": 1,
  "tool_name": "api_query"
}

Error Handling

Returns structured error responses for:

• Invalid endpoints or methods
• Authentication/authorization failures
• Malformed request data
• Server-side processing errors

Integration

This tool complements specialized workflow tools by providing comprehensive API access for:

• System monitoring and diagnostics
• Configuration verification
• Troubleshooting workflow issues
• Administrative operations

Use specialized Focused Tools like create_pipeline, delete_flow, add_pipeline_step, and configure_flow_step for mutation operations. api_query is strictly read-only for discovery and monitoring.