dadiaomengmeimei

dadiaomengmeimei / NL2AutoAPI

Public

NL2AutoAPI is an engineering system for structured-data Q&A: it distills table schemas, statistics, and historical samples into reusable API assets, then routes natural-language queries to stable SQL execution pipelines.

10
1
100% credibility
Found Mar 23, 2026 at 10 stars -- GitGems finds repos before they trend. Get early access to the next one.
Sign Up Free
AI Analysis
Python
AI Summary

NL2AutoAPI builds reusable question-answering APIs from database tables by pre-generating stable SQL pipelines with human review and runtime routing.

How It Works

1
🔍 Discover easy data questions

You find a tool that lets everyday people ask simple questions about their database tables, like 'How many employees in sales?' without writing code.

2
🔗 Connect your data source

Link your database so the tool understands your tables, fields, and sample values, making everything ready to learn from your real data.

3
🛠️ Build smart question answers

With one click, create a collection of ready-to-use answers for common questions based on your data patterns, like counts or lists.

4
✏️ Review and polish answers

Browse the question-answer pairs in a friendly screen, tweak descriptions or fixes if needed, and approve the best ones.

5
Start asking questions
✅
Perfect match

The tool picks the right answer and shows reliable results.

🔄
Needs tweak

If unclear, refine and improve for next time.

🎉 Reliable data insights

Now you chat with your database anytime, getting accurate, auditable answers that improve over time.

Sign up to see the full architecture

4 more

Sign Up Free

Star Growth

See how this repo grew from 10 to 10 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 NL2AutoAPI?

NL2AutoAPI is a Python engineering system that distills table schemas, statistics, and historical samples into reusable API assets for structured-data Q&A. It pre-builds stable SQL pipelines via CLI commands like `python main.py build`, then routes natural-language queries to execution paths for fast, auditable results. Users get a Gradio UI for review, exports to JSON or OpenAPI, and closed-loop feedback from runtime failures.

Why is it gaining traction?

It skips flaky on-the-fly LLM SQL generation by asset-ifying queries upfront—80% rule-based from data distributions, 20% LLM gap-filling—yielding replayable, iterable APIs. Runtime routing uses top-K recall plus LLM selection for reliable hits, with failures auto-escalating to review queues. Developers value the agent-style schema auto-fix loop that learns from real DB executions, plus diff-driven cascade updates without full rebuilds.

Who should use this?

Data engineers at mid-size firms building single-table Q&A over HR or ops databases, where multi-table joins are rare. Analytics teams needing auditable NL2SQL for dashboards, avoiding live hallucinations. Python shops prototyping internal tools that evolve via human feedback on historical queries.

Verdict

Solid foundation for reusable NL2AutoAPI assets and pipelines, but 10 stars and 1.0% credibility signal early-stage rawness—docs are README-focused, no broad tests. Try the CLI/UI flow on a demo DB if auditable Q&A fits; otherwise, watch or contribute.

(198 words)

Sign up to read the full AI review Sign Up Free

Similar repos coming soon.