withcoral

withcoral / coral

Public

One SQL interface over APIs, files, and live sources — built for agents.

withcoral.com
17
0
100% credibility
Found Apr 11, 2026 at 17 stars -- GitGems finds repos before they trend. Get early access to the next one.
Sign Up Free
AI Analysis
Rust
AI Summary

Coral is a local-first SQL runtime that unifies querying across APIs, files, and live data sources like GitHub and Slack, optimized for AI agent workflows.

How It Works

1
🔍 Discover Coral

You hear about Coral, a simple way to query all your company data like GitHub issues, Slack messages, or logs using everyday SQL—no more juggling separate tools.

2
📥 Install Easily

Grab Coral with a quick install command and see it ready on your computer in seconds.

3
✨ Connect Your Sources

Run the friendly onboard guide to link your first services like GitHub or Slack—it asks simple questions and sets everything up securely.

4
📋 Explore Tables

Check out the list of tables from your connected sources to see what's ready to query.

5
🔍 Run Your First Query

Type a SQL question like recent releases or team messages, and get clean results instantly right in your terminal.

6
Power Up with Agents
🧑‍💻
Query Yourself

Keep running SQL directly for quick insights anytime.

🚀
Connect to Agent

Launch a helper server so your AI can ask questions across all your data.

🎉 Unlock Cross-Source Magic

Now you get fast answers blending data from everywhere, saving hours of manual work—your data feels alive and easy!

Sign up to see the full architecture

5 more

Sign Up Free

Star Growth

See how this repo grew from 17 to 17 stars Sign Up Free
Repurpose This Repo

Repurpose is a Pro feature

Generate ready-to-use prompts for X threads, LinkedIn posts, blog posts, YouTube scripts, and more -- with full repo context baked in.

Unlock Repurpose
AI-Generated Review

What is coral?

Coral provides a local-first SQL runtime that unifies queries across APIs like GitHub and Slack, local files like JSONL or Parquet, and live sources—built specifically for AI agents. Install via Homebrew, run `coral onboard` to add sources with guided auth, then query with `coral sql "SELECT * FROM github.releases"`. It exposes the same runtime over MCP for agents, avoiding custom tool glue. Written in Rust, it leverages DataFusion for fast SQL execution.

Why is it gaining traction?

Unlike scattered API tools that force agents into multiple calls with repeated auth and pagination, Coral delivers cross-source SQL reasoning in one interface, cutting token use and brittleness. Developers dig the CLI for quick schema inspection via `coral.tables` and agent-ready MCP stdio server. Early bundled sources like GitHub and Slack make onboarding instant, standing out from verbose alternatives like github coral tpu or coral island github hacks.

Who should use this?

AI agent builders querying SaaS tools (GitHub issues, Slack messages) or files without stitching prompts. Teams tired of agent tool sprawl in workflows like incident response or code reviews. Devs prototyping local-first data agents, especially those eyeing coral ai github integrations over one-off scripts.

Verdict

Promising for agent devs, but at 17 stars and 1.0% credibility, it's pre-1.0—docs are solid via withcoral.com, but expect rough edges in source coverage. Try for GitHub/Slack proofs-of-concept; contribute if you build custom sources.

(198 words)

Sign up to read the full AI review Sign Up Free

Similar repos coming soon.