Cost Guard MCP MCP

v0.3.0

io.github.mcpsmiths/cost-guard-mcp

Pre-flight query cost and result-size guardrails for AI agents on BigQuery, Snowflake, Databricks

transport:stdio runtime:pypi

View source

Target client

run in your project directory

GOOGLE_APPLICATION_CREDENTIALS

SNOWFLAKE_ACCOUNT

SNOWFLAKE_USER

SNOWFLAKE_ROLE

SNOWFLAKE_PRIVATE_KEY_PATH

SNOWFLAKE_PRIVATE_KEY_PASSPHRASE

SNOWFLAKE_PASSWORD

DATABRICKS_SERVER_HOSTNAME

DATABRICKS_HTTP_PATH

DATABRICKS_TOKEN

DATABRICKS_CLIENT_ID

DATABRICKS_CLIENT_SECRET

Held in this page only — never stored, logged, or sent anywhere but back to your screen.

claude mcp add cost-guard-mcp -e GOOGLE_APPLICATION_CREDENTIALS=<GOOGLE_APPLICATION_CREDENTIALS> -e SNOWFLAKE_ACCOUNT=<SNOWFLAKE_ACCOUNT> -e SNOWFLAKE_USER=<SNOWFLAKE_USER> -e SNOWFLAKE_ROLE=<SNOWFLAKE_ROLE> -e SNOWFLAKE_PRIVATE_KEY_PATH=<SNOWFLAKE_PRIVATE_KEY_PATH> -e SNOWFLAKE_PRIVATE_KEY_PASSPHRASE=<SNOWFLAKE_PRIVATE_KEY_PASSPHRASE> -e SNOWFLAKE_PASSWORD=<SNOWFLAKE_PASSWORD> -e DATABRICKS_SERVER_HOSTNAME=<DATABRICKS_SERVER_HOSTNAME> -e DATABRICKS_HTTP_PATH=<DATABRICKS_HTTP_PATH> -e DATABRICKS_TOKEN=<DATABRICKS_TOKEN> -e DATABRICKS_CLIENT_ID=<DATABRICKS_CLIENT_ID> -e DATABRICKS_CLIENT_SECRET=<DATABRICKS_CLIENT_SECRET> -- uvx cost-guard-mcp