no-human-ai/no_human

From ticket to reviewed pull request. Free and open-source, on your machine.

What it solves

no_human is an AI coding factory that automates the entire software development lifecycle from ticket intake to the creation of a reviewed pull request. It eliminates the need for manual coding and reviewing for routine tasks by providing a trusted, autonomous loop that handles planning, implementation, testing, and adversarial review.

How it works

The system operates as a bounded loop with several strict gates to ensure code quality and security:

  • Planning: It creates a detailed plan with acceptance criteria based on the ticket and repository analysis before writing any code.
  • Adversarial Review: A separate AI model, which has not seen the coder's process, is tasked with refuting the "done" status and provides a pass/fail checklist citing specific files and lines.
  • Tamper Guard: The system mechanically counts deleted tests, new skips, or tautological assertions to prevent the AI from "cheating" the tests.
  • Reproduction Gate: For bug fixes, it verifies that the reproduction tests fail on the old code and pass on the new code.
  • Integration: It syncs tickets from Jira, Linear, or monday.com and opens PRs on GitHub or GitLab.
  • Human-in-the-loop: If the AI cannot finish a task, it parks the task with a specific question for the user rather than guessing.

Who it’s for

Software engineers and teams who want to automate the repetitive parts of their development workflow and integrate AI agents into their existing project management tools (Jira, Linear, etc.) and version control systems.

Highlights

  • Autonomous Loop: Moves tasks from backlog to reviewed PR without human intervention.
  • Adversarial Review: Uses an independent model to verify work, avoiding self-scoring.
  • Tamper Detection: Prevents the AI from bypassing tests or deleting them to pass.
  • MCP Server: Includes a Model Context Protocol server to allow integration with agents like Claude Code or Cursor.
  • Tracker Integration: Native support for Jira, Linear, and monday.com for ticket synchronization.

Related

  • Project
    microsoft/Build-your-own-copilot-Solution-Accelerator

    A solution accelerator providing reusable architecture and code to build custom enterprise copilots using Azure OpenAI and Azure AI Search for RAG-based data interaction.

  • Project
    holoviz/lumen

    An open-source agent-based framework for chatting with data and RAG, enabling users to generate data pipelines and visualizations via natural language.

  • Project
    shlokkhemani/rabbithole

    An infinite canvas for learning that allows users to branch into new documents by asking questions at any point in existing content.

  • Project
    0rangec3t/Black-cat

    A hypothesis-driven red teaming framework for Claude Code that replaces linear pentest pipelines with a state machine to automate complex penetration testing.