compforge/devloop

A guard-railed, closed-loop workflow for AI coding agents: live state bus + execution-level hard intercepts for Claude Code, Codex, and DeepSeek Harness (GitHub PR / GitLab MR). From step-level to requirement-level; eval-driven, spec-driven, human-in-the-loop.

This repository also configures its own agents. See what devloop tells them →

4Stars on the repository
27Mods indexed here, across every type
2d agoLast push, which is what freshness is scored on
noneNo LICENSE: all rights reserved, so bodies are not copied

code-taste

01

compforge/devloop

Plugin Claude Code

Bundles 1 skill · 64 tokens together

Engineering judgment for architecture, boundaries, naming, maintainability, documentation, and code review.

not rated 4 2d ago A tokens not measured

devloop

02

compforge/devloop

Plugin Claude Code

Bundles 7 skills, 3 commands, 9 hooks · 513 tokens together

TypeScript dev-loop runtime with shared Board and policy: git/PR (GitHub + GitLab), cwd-aware context, lint/test gates, and thin Claude Code hooks.

not rated 4 changed 2d ago A tokens not measured

quality

03

compforge/devloop

Plugin Claude Code

Bundles 3 skills · 266 tokens together

Evidence-driven skills that operate project-owned E2E, performance, and agent trajectory quality capabilities.

not rated 4 changed 6d ago A tokens not measured