hardness1020

hardness1020 / Leeway

Public

A workflow-driven AI agent framework that executes YAML-defined decision trees. Bridging rigid automation and unpredictable agents

12
4
100% credibility
Found Apr 11, 2026 at 12 stars -- GitGems finds repos before they trend. Get early access to the next one.
Sign Up Free
AI Analysis
Python
AI Summary

Leeway is a workflow-driven AI agent framework that lets users define decision trees in YAML for reliable, auditable execution of software engineering tasks using LLMs and built-in tools.

How It Works

1
👀 Discover Leeway

You find this helpful AI assistant on GitHub that runs smart workflows for checking code health and more.

2
🚀 Get it running

Follow simple steps to install and open the colorful terminal interface on your computer.

3
🧠 Link your AI brain

Connect a smart AI service like Claude so your assistant can think and act on its own.

4
📋 Pick a ready workflow

Choose the built-in code-health check or type your own question to get started right away.

5
✨ Watch it work

See your assistant scan files, analyze issues, and run checks automatically in real-time.

✅ Enjoy perfect results

Receive a clear summary of your project's health with tips to improve it effortlessly.

Sign up to see the full architecture

4 more

Sign Up Free

Star Growth

See how this repo grew from 12 to 12 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 Leeway?

Leeway is a Python framework for AI agents that execute YAML-defined decision trees, bridging rigid automation and unpredictable agents. You design workflows with branches, loops, parallel execution, and signals, while AI handles flexible reasoning inside each node using 21 built-in tools for file ops, shell commands, web search, cron jobs, memory, and subagents. It offers a React terminal UI, leeway api for providers like Anthropic or OpenAI, and scheduling via cron daemon.

Why is it gaining traction?

Unlike pure agents that hallucinate paths or rigid scripts lacking smarts, Leeway delivers deterministic transitions with AI leeway—meaning controlled flexibility where humans spec the graph and models execute steps. Standout hooks include human-in-the-loop gates, progressive skills for on-demand knowledge, MCP integration, and spec-driven workflows on GitHub that run reliably with 190 passing tests. Devs dig the CLI commands like /workflows or /skills for quick iteration.

Who should use this?

Python devs automating code reviews, repo health checks, or scheduled tasks—think backend engineers tired of flaky agents, ops folks scripting audits, or teams needing decision-tree agents for file analysis without chaos. Ideal for those blending automation with AI in controlled environments like CI or local dev.

Verdict

Worth a spin for structured agent workflows; uv install, solid docs, and CLI make it approachable despite 12 stars and 1.0% credibility signaling early maturity. Strong tests and extensibility outweigh the low traction—prototype your next automation here.

(198 words)

Sign up to read the full AI review Sign Up Free

Similar repos coming soon.