openapi: 3.1.0
info:
title: Dewey MCP Feeds Health API
version: '0.1'
summary: FastMCP server exposing read-only search over the Inquirer archive.
description: Dewey MCP is a containerized FastMCP server for read-only search over a news archive backed by Azure AI Search. It exposes a single MCP tool, `search_archive`, that accepts a search query plus optional date and author filters and returns ranked article chunks with metadata. The server is open source under `phillymedia/dewey-mcp` and is the protocol surface for the broader Dewey AI librarian project (Lenfest Institute fellowship, supported by Microsoft and OpenAI). This OpenAPI describes the server's HTTP envelope (MCP protocol endpoint, liveness, and readiness probes); the MCP tool itself is described in `components.schemas`.
contact:
name: Dewey MCP
url: https://github.com/phillymedia/dewey-mcp
servers:
- url: http://127.0.0.1:8000
description: Default local server (configurable via MCP_HOST and MCP_PORT)
tags:
- name: Health
description: Liveness and readiness probes.
paths:
/livez:
get:
operationId: getLiveness
summary: Get Liveness Probe
description: Returns process liveness status.
tags:
- Health
responses:
'200':
description: Process is alive.
content:
application/json:
schema:
type: object
properties:
status:
type: string
example: ok
/readyz:
get:
operationId: getReadiness
summary: Get Readiness Probe
description: Returns readiness status; ready when the configured Azure AI Search index is reachable.
tags:
- Health
responses:
'200':
description: Service is ready to serve search traffic.
content:
application/json:
schema:
type: object
properties:
status:
type: string
example: ready
'503':
description: Service is not ready (Azure AI Search unreachable).