wesleysimplicio

wesleysimplicio / simplicio-dev-cli

Public

Your tasks with 99% accuracy using any LLM (Claude, DeepSeek, Codex, Gemini, Hermes, OpenClaw, Cursor).

pypi.orgprojectsimplicio-cli agentic-ai ai-agent ai-tools anthropic claude
11
1
89% credibility
Found May 28, 2026 at 11 stars -- GitGems finds repos before they trend. Get early access to the next one.
Sign Up Free
AI Analysis
Python
AI Summary

This is a developer tool that acts as a smarter middle layer between you and your AI coding assistant. When you ask the AI to make a code change, the tool wraps your request in a structured blueprint built from your project's actual files and past patterns. It also generates tests alongside the code, runs them, and automatically retries if something fails. The result: AI coding assistants go from getting it wrong most of the time to getting it right almost every time, on any AI model — including free local ones running on your own machine. The tool installs in one command, auto-activates inside Claude Code on first use, and requires no API key if you already have a subscription.

How It Works

1
đź’ˇ You hear about a smarter way to work with AI coding helpers

A developer friend tells you there's a way to make AI coding assistants dramatically more accurate — from guessing wrong most of the time to getting it right nearly every time, without changing which AI you use.

2
📦 You install it in one line

You run one simple command and the tool is ready. On your very first use, it quietly sets itself up inside your existing AI assistant so every coding request you make gets a smarter, structured approach.

3
🏗️ You describe what you want to build

Instead of hoping the AI guesses correctly, you simply type what you need — like "hide the Delete button for non-admins in the admin screen" — and the tool hands the AI a precise, layered blueprint of your project.

4
🔬 The AI generates code, tests, and proof — all at once

The AI produces a complete plan: the exact code change to make, test cases that prove it works for both the allowed and blocked scenarios, and a way to verify it visually.

5
The tool checks your project's tests automatically
âś…
Tests pass on the first try

Everything works right away — the AI nailed it.

🔄
Tests fail — AI learns from the error and retries

The AI sees the failure, understands what went wrong, and tries a corrected version automatically.

🎉 Your code change is verified and applied

The exact change you asked for is in your project, tested, and confirmed working — without you having to write it yourself.

Sign up to see the full architecture

4 more

Sign Up Free

Star Growth

See how this repo grew from 11 to 11 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 simplicio-dev-cli?

simplicio-dev-cli is a Python-based task automation tool that takes a simple one-line request like "hide the Delete button for non-admins" and transforms it into a verified code change with diff, tests, and a pass/fail loop. It works with any LLM provider—Claude, DeepSeek, Codex, Gemini, OpenRouter, or even local Ollama models—without requiring you to change your existing workflow. The core idea is a six-layer prompt contract that gives the model exactly what it needs: target file, success criteria, constraints, and an exact output format. The CLI then applies the diff, runs your test suite, and retries up to three times if verification fails.

Why is it gaining traction?

The benchmarks are the hook. Across seventeen models tested—from tiny 1B-parameter models running locally to frontier 2026 models—every single one gained at least 14 percentage points when wrapped in simplicio's contract. Frontier models hit 99-100% pass rates. The most striking number: a 1.5B Qwen model running on a four-year-old laptop with zero network access reached 88% pass rate versus 32% without the wrapper. For teams already paying for Claude Code Pro or Codex CLI subscriptions, the tool auto-installs a skill and hook that routes code-edit prompts through the contract at no extra cost—using your existing login, no API key required.

Who should use this?

Developers using Claude Code, Codex CLI, or any OpenAI-compatible LLM for code edits will see the most immediate benefit. It's particularly useful for teams with existing test suites who want LLM-generated changes automatically verified before landing. Teams running local models on constrained hardware will appreciate the offline Ollama support. If you're already happy with how your LLM handles code tasks, this won't change your mind—but if you're seeing hallucinated targets, missing diffs, or one-shot failures, the contract wrapper is worth trying.

Verdict

With an 0.8999999761581421% credibility score and only 11 stars, this is a young project with a compelling proof-of-concept and unusually thorough benchmarking methodology. The numbers are reproducible and the architecture is clean. Try it via `pip install simplicio-cli` if you have Claude Code or an OpenRouter key. The auto-activation for Claude Code users is frictionless—just install and type your task as normal.

Sign up to read the full AI review Sign Up Free

Similar repos coming soon.