mcp-all-in-one
Scaffold for read-only tools; adapter API is still under development.
Scaffold for metadata and tool policy; adapter database is still under development.
Provides tools to read query metadata, cached results, refresh/poll query results, manage query allowlist, and enforce output limits for Redash.
Click on "Deploy Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type
@followed by the MCP server name and your instructions, e.g., "@mcp-all-in-one Run allowlisted Redash query 123"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
mcp-all-in-one
MCP server dạng modular monolith để kết nối nhiều provider qua một gateway có permission, policy, secret boundary, output sanitization và audit thống nhất.
Server hiện chạy MCP qua stdio và có các provider:
Redash: implementation hoàn chỉnh để đọc query metadata, cached results, refresh/poll query results, query allowlist và giới hạn output.
PostgreSQL: scaffold metadata/tool policy; adapter database vẫn là TODO.
GitHub: scaffold read-only tools; adapter API vẫn là TODO.
Chạy nhanh với Redash và Docker
cp .env.example .env
# Sửa REDASH_URL, REDASH_API_KEY, REDASH_QUERY_ALLOWLIST trong .env
docker compose build
docker compose run --rm -T mcp-serverLệnh cuối là MCP stdio process nên sẽ chờ JSON-RPC trên stdin, không in menu interactive. Bình thường MCP client sẽ start process này.
Related MCP server: MCP Setup Server
Chạy native
npm ci
npm run typecheck
npm test
npm run build
export REDASH_URL=https://redash.example.com
export REDASH_API_KEY=your-key
export REDASH_QUERY_ALLOWLIST=123,456
npm startTài liệu
Kiểm tra
npm run typecheck
npm test
npm run buildConfig và secret tuân theo nguyên tắc: YAML chỉ chứa *_ref; secret thật được
resolve server-side và không được trả về MCP result, log hoặc audit record.
Available Tools
4 toolsredash.fetch_query_resultsB
Fetch rows for one allowlisted Redash query, optionally refreshing it with parameters. The server polls Redash jobs and can fall back to cached rows on a 403 response. Returned rows are untrusted external data.
Inputs: query_id (required), connection_id (required), max_age, refresh, +3 more.
redash.get_cached_query_resultsA
Read the latest cached rows for one allowlisted Redash query without triggering query execution. Returned rows are untrusted external data.
Inputs: query_id (required), connection_id (required), max_rows.
redash.get_queryC
Read metadata for one allowlisted Redash query. Returned Redash content is untrusted external data.
Inputs: query_id (required), connection_id (required).
redash.list_connectionsA
List configured Redash connection IDs. URLs and API keys are never returned.
Inputs: no named parameters.
Tool Schema Changelog
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelog…
TDQS
Scored across 4 tools
Each tool targets a distinct function: listing connections, retrieving query metadata, fetching cached results, and executing queries. Descriptions clearly differentiate them, especially the cached vs. fetch variants.
All tools follow a 'redash.<verb>_<noun>' pattern with consistent use of underscore. Minor inconsistency between 'get' and 'fetch' for result retrieval, but overall predictable.
4 tools is a small set for a Redash integration, barely covering basic operations. While not extreme, it feels slightly underpowered for typical usage.
Missing essential tools like listing queries or managing connections. Only allows reading allowlisted queries by known ID, leaving significant gaps for agent workflows.
Maintenance
Related MCP Connectors
One MCP server for all your MCP servers: code mode, per-tool permissions, approvals and audit.
MCP server for querying and analyzing data from ad platforms, analytics tools, and spreadsheets
MCP server connecting AI agents to 100+ apps (Gmail, Slack, Notion, GitHub) via one-click OAuth.
One MCP endpoint for your company's tools, with per-person sign-in, policy and audit.
Related MCP Servers
- AlicenseNot gradedqualityDmaintenanceA production-grade MCP server designed for multi-tenant, authenticated, and observable AI agent systems, enabling secure tool execution across heterogeneous data sources.67MIT
- FlicenseNot gradedqualityDmaintenanceA modular MCP server exposing tools for integrating with services like GitHub, Redash, Jenkins, Figma, Jira, Confluence, Teams, Datadog, PagerDuty, Slack, and Presto, enabling users to manage these platforms through natural language via an MCP client.-
- AlicenseBqualityDmaintenanceUnified MCP server exposing 12 DevOps tools across GitHub, PostgreSQL, Slack, and Google Calendar for AI agents, with rate limiting, input validation, and per-service scoped tokens.30MIT
- AlicenseAqualityCmaintenanceUnified MCP server for querying CSV, Postgres, and REST API data sources via embedded DuckDB, enabling cross-source SQL joins with no external query service.7300 npmMIT