Skip to main content
AdOpenFree logoPromote your productReach more potential users and drive product growth and revenue.Advertise
Favicon of LoopForge

LoopForge

Free Listing

LoopForge is an open-source, self-hosted platform for AI software delivery. It orchestrates coding-agent workflows with cloud sandboxes, scripts and human approval gates, and offers Local workflows for coding assistants.

Visit Tencent/LoopForge

Overview

LoopForge is a self-hosted workspace where teams ship software with coding agents. A Main Bot can open Docker sandboxes, check out repositories, delegate to project roles, and open PRs; repeatable SOPs define agents, scripts and human gates so delivery follows the same reviewed path.

Key Features

  • Main Bot chat workbench with per-conversation executor, model and permission mode; roles, skills, requirements, SOPs and repositories can be referenced with @.
  • Conversation workspaces use on-demand Docker sandboxes with repositories checked out and project roles as subagents; runs can be paused, resumed or taken over.
  • Visual SOP designer supports Agent, Script, Human gate, Sub-SOP and Preview nodes, rework-on-failure edges, validation, version history and publishing.
  • Hard gates use build, unit test and lint exit codes; failed checks attach logs and send work back for automatic rework.
  • Scheduled GitHub Issue sync and issue autopilot can pull issues, run fixes in parallel and resolve merge conflicts on the integration branch.
  • Post-run review scores runs and generates one-click updates for SOP steps, role guides and quality gates.

Use Cases

  • Teams that want repeatable delivery workflows with fixed stages and human sign-off can run SOPs instead of relying on ad-hoc agent chats.
  • Developers can use Main Bot chat for exploration, quick fixes and iterative work in a Docker sandbox with project repositories.
  • Product and engineering teams can track requirements from request to PR/MR, whether work is delivered by chat or an SOP.
  • Teams can schedule GitHub Issue sync and run routine bug fixes in the background with human review.

Getting Started

  • Install Git and clone the repository, then run ./start.sh.
  • Open the printed address and create the first account, which becomes the administrator. The service listens on 0.0.0.0:8090 by default.
  • Configure model API keys and GitHub authorization after startup. For LoopForge Local, run npx -y loopforge-cli@latest install claude inside the target project, then use /start-devflow.

Deployment & Requirements

  • Platform edition requires Docker with Compose and Python 3, plus Git to clone the repository.
  • LoopForge Local requires Node.js 20+ and Python 3.8+ with a supported coding assistant.
  • Package and image downloads require internet access; first-use installs of some executors need npm or provider access.

Before You Adopt

  • License: Other. Review its terms before using, modifying, or distributing the project.
  • The default Compose setup uses Docker socket access and is intended for trusted teams, not untrusted multi-tenant isolation.
  • New installations listen on all interfaces and are designed for a trusted team, with signed-in users able to read project data across the instance.
  • Package and image downloads depend on internet access.
  • The management pages support English and Chinese, while the chat workbench is currently Chinese only.

Comments

Sign in to leave a comment.

More like LoopForge

Favicon of Elephant-Goldfish Model

Elephant-Goldfish Model

Free ListingStars: 18.6K

Design with context, review with fresh eyes

Coding Agents

EGM is an open-source plugin for AI assistants that helps teams draft plans with full project context, then have fresh reviewers who did not see the conversation identify missing or unclear details before implementation.

Favicon of TokenEconomizer

TokenEconomizer

Free ListingStars: 5.9K

Experience-driven cost control for coding agents

Coding Agents

TokenEconomizer is a plugin that learns from prior coding-agent sessions in a repository to reduce costs by remembering repository facts, pruning unused tools, and intervening in failure loops without changing the agent.

Favicon of Qwen Code

Qwen Code

Free ListingStars: 29.6K

Open-source AI coding agent in your terminal

Coding Agents

Qwen Code is an open-source AI coding agent that lives in your terminal, helping turn ideas into code with agentic workflows and extensible development tools.