Keyword Research MCP
Enables retrieval and normalization of Google Ads keyword-planning evidence.
README
Keyword Research MCP
Keyword Research MCP is a lightweight local MCP server for retrieving and normalizing Google Ads keyword-planning evidence. It is under active development; the current repository contains the executable project foundation plus the typed, tested research interface and production Google Ads adapter. MCP tool registration described in the implementation plan is still being built.
Paid Competition in Google Ads describes advertiser activity. This project does not present it as organic ranking difficulty and does not calculate a composite keyword score.
Requirements
- Python 3.10 through 3.14
- uv
Google Ads credentials are not needed for tests. They are required when the production adapter connects to Google Ads.
Google Ads configuration
The production adapter uses google-ads==31.2.0 and explicitly selects Google
Ads API v25. Configure exactly one target customer and one authentication
method. Customer IDs may contain hyphens; they are normalized in memory.
For user OAuth, set:
export GOOGLE_ADS_DEVELOPER_TOKEN="developer-token-placeholder"
export GOOGLE_ADS_CUSTOMER_ID="123-456-7890"
export GOOGLE_ADS_LOGIN_CUSTOMER_ID="987-654-3210" # optional manager account
export GOOGLE_ADS_CLIENT_ID="client-id-placeholder"
export GOOGLE_ADS_CLIENT_SECRET="client-secret-placeholder"
export GOOGLE_ADS_REFRESH_TOKEN="refresh-token-placeholder"
For a service account, omit the three user-OAuth settings and set:
export GOOGLE_ADS_DEVELOPER_TOKEN="developer-token-placeholder"
export GOOGLE_ADS_CUSTOMER_ID="123-456-7890"
export GOOGLE_ADS_JSON_KEY_FILE_PATH="/path/to/service-account.json"
export GOOGLE_ADS_IMPERSONATED_EMAIL="ads-user@example.test" # optional
Alternatively, point GOOGLE_ADS_CONFIGURATION_FILE_PATH at a local
google-ads.yaml:
developer_token: developer-token-placeholder
customer_id: "123-456-7890"
login_customer_id: "987-654-3210"
client_id: client-id-placeholder
client_secret: client-secret-placeholder
refresh_token: refresh-token-placeholder
Optional process settings are KEYWORD_RESEARCH_CACHE_CAPACITY (default 128)
and KEYWORD_RESEARCH_LOG_LEVEL (default INFO). Credential files,
google-ads.yaml, and .env files are ignored by Git.
Set up and run
uv sync
uv run keyword-research-mcp
The command runs over stdio, so it waits silently for an MCP client and writes
protocol messages only to standard output.
Development
Run the same quality gate used in CI:
uv run ruff format --check .
uv run ruff check .
uv run pyright
uv run pytest
uv build
See the implementation plan for the intended tool contracts, architecture, delivery phases, and product boundaries.
License
MIT
Recommended Servers
playwright-mcp
A Model Context Protocol server that enables LLMs to interact with web pages through structured accessibility snapshots without requiring vision models or screenshots.
Magic Component Platform (MCP)
An AI-powered tool that generates modern UI components from natural language descriptions, integrating with popular IDEs to streamline UI development workflow.
Audiense Insights MCP Server
Enables interaction with Audiense Insights accounts via the Model Context Protocol, facilitating the extraction and analysis of marketing insights and audience data including demographics, behavior, and influencer engagement.
VeyraX MCP
Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.
graphlit-mcp-server
The Model Context Protocol (MCP) Server enables integration between MCP clients and the Graphlit service. Ingest anything from Slack to Gmail to podcast feeds, in addition to web crawling, into a Graphlit project - and then retrieve relevant contents from the MCP client.
Kagi MCP Server
An MCP server that integrates Kagi search capabilities with Claude AI, enabling Claude to perform real-time web searches when answering questions that require up-to-date information.
E2B
Using MCP to run code via e2b.
Neon Database
MCP server for interacting with Neon Management API and databases
Exa Search
A Model Context Protocol (MCP) server lets AI assistants like Claude use the Exa AI Search API for web searches. This setup allows AI models to get real-time web information in a safe and controlled way.
Qdrant Server
This repository is an example of how to create a MCP server for Qdrant, a vector search engine.