EricTechPro/super-board

Autonomous GitHub Project build → QA → review loop for Claude Code. One-line install.

What it solves

super-board provides an automated, agentic workflow for software development. It allows developers to add tasks to a board and have an AI agent system autonomously handle the build, testing, and review processes, ultimately merging pull requests once they pass all quality gates.

How it works

The system functions as a set of skills integrated into an AI coding environment (such as Claude Code). It uses a "board" metaphor where tasks are added to a backlog. The system then routes these tasks through three automated lanes: a builder that creates changes and drafts PRs, a tester that verifies evidence and checks for gaps, and a reviewer that validates the work against project policies before merging. It includes safety "guard hooks" that prevent dangerous operations like unauthorized file deletion or secret exposure.

Who it’s for

This project is for developers who use AI coding assistants and want to automate the repetitive parts of the software development lifecycle—specifically planning, building, testing, and reviewing code changes—to maintain a "self-improving" software factory.

Highlights

  • Automated Lanes: Dedicated AI-driven lanes for building, QA, and reviewing code.
  • Safety Guards: Built-in hooks to prevent common AI mistakes like leaking secrets or modifying files outside the project scope.
  • Model Routing: Automatically selects different AI models (e.g., Haiku, Sonnet, Opus) based on the complexity of the task.
  • Standardized Workflow: Enforces a consistent writing standard for commits, tickets, and PRs across projects.
  • Extensible Skills: Modular design allows users to add custom skills to the board's capabilities.

Related

  • Project
    ruvnet/open-claude-code

    An open-source, clean-room implementation of the Claude Code CLI that provides an agentic terminal interface for coding tasks, supporting custom tools, MCP integration, and multiple AI providers.

  • Project
    marcoaapfortes/Mantic.sh

    A context-aware code search engine that prioritizes relevance over raw speed to provide AI agents and developers with precise, intent-based file retrieval.

  • Project
    microsoft/RPG-ZeroRepo

    RPG-ZeroRepo provides a Repository Planning Graph (RPG) that acts as a persistent control layer for AI coding agents, helping them plan, generate, understand, and update codebases across long tasks. It includes CoderMind for agent integration, ZeroRepo for generating repositories from requirements, RPG-Encoder for mapping codebases back to RPGs, and RepoCraft for evaluation.

  • Project
    sting8k/pi-vcc

    pi‑vcc is a deterministic, algorithm‑only conversation compactor for the Pi coding‑assistant. It replaces Pi’s LLM‑based summariser with fast extraction of goals, file changes, commits, commands, open issues, and user preferences, reducing session size by ~98 % with millisecond latency and no token cost. The plugin works automatically or via `/pi‑vcc` commands, keeps the compacted history searchable through `vcc_recall`, and is configurable via a JSON file. MIT‑licensed.