Devloops

Devloops: Reusable agent-loop skills for coding agents that handle routine maintenance tasks in adversarial loops.

devloops provides reusable agent-loop skills for coding agents using a producer, worker, and gatekeeper structure. The producer audits code and files candidate issues, the worker implements approved issues as pull requests, and the gatekeeper acts as an independent reviewer to find defects. The worker and gatekeeper interact repeatedly until the gatekeeper approves the pull request, while humans retain control over merging and key decisions. The system integrates via a configuration file and helper scripts installed in the project directory, relying on Python and the GitHub CLI. It is built for developers who want to automate routine maintenance work safely using AI coding agents.

Devloops at a glance

Best for: Software engineer
Runs on: CLI, Python, GitHub
Pricing: Free
Released: November 2026
Category: AI coding tool
Website: github.com/fritzprix/devloops

Key features

Producer audits code and files candidate issues
Worker implements approved issues as pull requests
Gatekeeper reviews pull requests independently
Tick gate schedules and optimizes loop execution
Configurable task periods and limits
Safety rules prevent agents from merging code

Use it for

Automating routine code maintenance tasks
Auditing codebases for candidate issues
Reviewing pull requests independently
Managing adversarial agent loops for coding

Alternatives to Devloops

AgentOps — Track token usage, costs, and traces across 400 plus LLMs and agent frameworks.
Loop Engineering — Design agent loops to discover work and verify results automatically.
Martin Loop — Enforces budgets, stop rules, and verifiers for coding agents.
Loopy — Provides repeatable prompts with checks, proof, and stopping conditions.
pstack — Improves AI coding agent outcomes with rigorous workflows.
Agent Skills — Encodes senior-engineer workflows and quality gates into 24 skills.
jevgrep — Reduces coding agent costs by providing targeted, relevant repository context.
Agentloop — Provides a clean agent loop with provider-agnostic LLM access, hooks, and safety budgets.
Lever — Agents iterate autonomously against set metrics and win criteria for hours.
Archon — Orchestrate multiple AI agents with isolated workflows via a single command layer.
BlastLoop — It applies an operational structure to agent coding to prevent rapid failures.
Agent Beacon — Retains coding agent knowledge across tools so errors do not repeat.

All Devloops alternatives