coral
One SQL interface over APIs, files, and live sources — built for agents.
Coral is an MCP server that provides AI agents with a single SQL query interface for accessing data from APIs, files, and live sources. Instead of making multiple tool calls to individual MCP servers or API wrappers, agents can write SQL queries that Coral translates into API calls and file reads, returning unified tabular results. It supports cross-source joins, handles pagination and authentication locally, and bundles integrations for GitHub, Linear, Slack, Datadog, Sentry, Stripe, and more. Benchmarks show agents using Coral are 20% more accurate and 2x more cost-efficient than using direct provider MCPs.
Key Features
Use Cases
- 01Query GitHub issues and pull requests alongside Linear tickets to track development progress
- 02Correlate Datadog monitoring metrics with Sentry error logs for debugging
- 03Search Slack conversations and cross-reference with GitHub commits for team context
- 04Aggregate Stripe payment data with customer information from multiple systems
- 05Build coding agents that query project metadata from files and APIs in one workflow
- 06Create dashboards by joining data from local JSONL/Parquet files with live API sources
Related MCP Servers
View moremarkitdown
Python tool for converting files and office documents to Markdown.
firecrawl
The context API to search, scrape, and interact with the web at scale. 🔥
prompts.chat
f.k.a. Awesome ChatGPT Prompts. Share, discover, and collect prompts from the community. Free and open source — self-host for your organization with complete privacy.
langflow
Langflow is a powerful tool for building and deploying AI-powered agents and workflows.
coral — FAQ
What is the Coral MCP server?+
Coral is an MCP server that gives AI agents a unified SQL interface for querying multiple data sources, APIs, and files. It translates SQL queries into API calls and returns tabular results, reducing the number of tool calls agents need to make.
How do I install and connect Coral to my AI agent?+
Install Coral via Homebrew on macOS (brew install withcoral/tap/coral), the install script on Linux, or download the desktop app. Then connect it to your agent with 'claude mcp add --scope user coral -- coral mcp-stdio' for Claude Code or similar commands for other clients. For Cursor and VS Code, manually add the MCP server configuration pointing to the coral mcp-stdio command.
Which AI clients work with Coral?+
Coral works with Claude Code, Codex, Cursor, VS Code, Claude Desktop, and OpenCode. Any client that supports the MCP protocol over stdio can connect to Coral's built-in MCP server.
Do I need API keys to use Coral?+
Yes, you need API keys or authentication tokens for each data source you want to query. Run 'coral source add --interactive <source>' to be prompted for credentials, which are stored locally. Examples include GitHub personal access tokens, Linear API keys, or Slack OAuth tokens.
Is Coral free and open source?+
Yes, Coral is licensed under Apache License 2.0 and available as open source on GitHub. All processing happens locally on your machine.
Can I add custom data sources to Coral?+
Yes, you can write custom source specs as YAML files that declare how to reach any HTTP API or local dataset. The repository includes community-contributed source specs, or you can create your own following the documentation.
How do I install coral?+
Open the source repository on GitHub and follow its README. coral is a mcp server — MCP Agents Market links you directly to the official repo.
Is coral free?+
coral is an open-source project hosted on GitHub. Check the repository for its license and any usage requirements.