API & DB Servers

Lower-level infrastructure tools for agencies and advanced developers managing the Axolot ecosystem at scale.

While the Master Server handles high-level content, the API and DB Servers provide raw access to the Axolot engine. These tools are designed for infrastructure orchestration and high-level technical tasks.

API Control Server

The API MCP is a dynamic bridge that allows the AI to discover and call any endpoint in the Axolot system, even those not yet documented or part of a specific module.

get_api_spec

Returns the full OpenAPI (Swagger) definition. This allows the AI to learn every available endpoint on the fly.

call_endpoint

Executes any HTTP request (GET, POST, PATCH) against the API. Used for creating bulk sites or updating system-wide configs.

Database Direct Server

For tasks where the API is too restrictive, the DB Server allows for Direct SQL Execution. This is critical for data recovery, bulk migrations, or complex reporting.

list_tables: Scans the entire PostgreSQL instance, including the isolated tenant schemas (site_*).
get_table_info(table): Inspects column types and constraints. The AI uses this to understand the underlying data model before writing SQL.
db_query(sql): Executes raw SQL. Use with caution. The AI will typically use this to perform complex joins or filter content across multiple sites.

Security Warning: These tools have "God Mode" access. Ensure your AXOLOT_API_TOKEN is restricted to Agency-level roles if you expose these servers to your AI.

AI-Ready Context

Raw SQL Example

// DB tool call from AI
{
  tool: "db_query",
  arguments: {
    query: "SELECT * FROM site_demo.blog_posts 
            WHERE status = 'published' 
            ORDER BY created_at DESC;"
  }
}

System Capability

"By combining get_api_spec and db_query, I can perform cross-site content audits and ensure every client instance follows the same data schema rules."

Logic & constraints

Use Slot keys as primary keys.

Do not hardcode content strings.

Respect TypeScript interfaces.