惯性聚合 高效追踪和阅读你感兴趣的博客、新闻、科技资讯
阅读原文 在惯性聚合中打开

推荐订阅源

P
Proofpoint News Feed
V
V2EX
WordPress大学
WordPress大学
Google DeepMind News
Google DeepMind News
Martin Fowler
Martin Fowler
小众软件
小众软件
Blog — PlanetScale
Blog — PlanetScale
月光博客
月光博客
The Cloudflare Blog
T
Tailwind CSS Blog
H
Help Net Security
腾讯CDC
爱范儿
爱范儿
人人都是产品经理
人人都是产品经理
H
Hackread – Cybersecurity News, Data Breaches, AI and More
The GitHub Blog
The GitHub Blog
Microsoft Security Blog
Microsoft Security Blog
Stack Overflow Blog
Stack Overflow Blog
D
DataBreaches.Net
C
Check Point Blog
量子位
酷 壳 – CoolShell
酷 壳 – CoolShell
美团技术团队
让小产品的独立变现更简单 - ezindie.com
让小产品的独立变现更简单 - ezindie.com

Show HN

GitHub - astefanutti/shaderbang: Shebang for Shaders Show HN: Generate Claude Code Workflows using Spec Driven Development approach Show HN: AI agents for UK GDAD PCF roles and their skills The Two Pillars: Mixer Mode and Meta-Software in the Reorganization of Software Work After AI GitHub - JaiCode08/teleport-env What 1,000+ Harness Experiments Taught Me About Self-Improving Agents Show HN: Liiists, a Markdown-first, iOS and CLI list app SwiperTab – Get this Extension for 🦊 Firefox (en-US) GitHub - kouhxp/fftext: Summarize, explain, fact-check, or translate any text, URL, or file. No GPU. No cloud. One command GitHub - sweetpad-dev/sweetpad: Develop Swift/iOS projects using VSCode GitHub - dogmaticdev/IRON: IRON a.k.a. Intermediate Representation Object Notation is a Interpreter/Database that is used to create Programming Languages. GitHub - sjhalani7/vaen: Package your AI coding harness into a portable .agent file, and share it across repos, teams, & the community without ever having to copy-paste instructions, skills, MCP config, or secrets. Show HN: Gandalf the Grader Show HN: Citadeld – replay any CI failure locally from a single file GitHub - tdortman/cuSBF: High-Performance GPU Super Bloom Filter coral-ai/claude-code-token-xray at main · Coral-Bricks-AI/coral-ai GitHub - ulyssestenn/funes: Funes is a Git-based framework for LLM-managed knowledge work: an AI Librarian ingests raw sources, builds an interlinked Markdown knowledge base, and uses it to produce cited reports, analyses, and other outputs. GitHub - ThatXliner/gah: Git Add Hunk, built for agents to use GitHub - harmont-dev/harmont-cli: Command-line client for the Harmont CI platform GitHub - brooksmcmillin/mcp-authflow: OAuth 2.0 Authorization Server framework for MCP servers GitHub - javaid-codes/audit-supply-chain-agents GitHub - amorey/gochan: A small library of common channel architectures for Go, inspired by Rust GitHub - arifozgun/OpenGem: Free, Open-Source AI API Gateway with Gemini, OpenAI & Anthropic Compatibility in 1 file GitHub - Pranesh950/BioPetals: 🌸 Run BIOxAI models at home, BitTorrent-style. Fine-tuning and inference up to 10x faster than offloading GitHub - cnguyen14/bounty-doctor: Diagnose a GitHub bounty issue before you waste hours: detects honeypot scam repos, AI-bot attempt swarms, and stale contests. Show HN: CoreMCP – MCP Server for On-Prem DBs Show HN: KittyHTML – Render HTML/CSS as an inline image in your terminal GitHub - bingud/filemat: Web-based file manager Show HN: TruthLens – Free multi-signal deepfake image detector GitHub - apexlocal-jz/claude-usage-tray: Windows system-tray app showing your Claude Code rate-limit usage at a glance. Zero deps, ~300 lines of PowerShell. Cross-IDE (works regardless of VS Code, Cursor, plain terminal).
goloop — the agentic loop CLI
mantyx · 2026-06-12 · via Show HN

// open source · MIT · single static binary

The agentic loop,
on the command line.

A supervisor model plans every iteration. A worker — Cursor or Claude Code — does the edits. State lives in plain markdown you can read, diff, and trust. Point it at a goal; walk away.

curl -fsSL https://raw.githubusercontent.com/mantyx-io/goloop/main/scripts/install.sh | bash

supervisor plans the move, emits a JSON action

worker edits the repo via Cursor / Claude Code

checkpoint writes .goloop/checkpoint.md

// 01 — what's inside

Built for autonomous coding,
not for a demo.

01

Supervisor / worker split

The planner never touches files. A dedicated worker does every edit — clean separation, predictable runs.

02

Pluggable backends

Supervise with ChatGPT, OpenAI, or Anthropic. Execute with Cursor CLI or Claude Code. Swap freely.

03

State you can read

No hidden database. Progress, blockers, and the iteration log are markdown files you can diff and edit.

04

Self-extending tools

The loop can author its own supervisor tools, exit 75, and auto-restart to pick them up.

05

Human in the loop

When only you can answer, it stops and asks with ask_user instead of hallucinating a decision.

06

Honest terminal UI

Streamed worker reasoning, a live iteration log, and full-screen wizards for setup — no spinners that lie.

07

Layered config

Models & auth once, globally. The goal per project. Flags override anything for a single run.

08

One static binary

Written in Go. No runtime, no node_modules. curl | bash and you're working.

// 02 — how it works

Read checkpoint. Decide. Act.

  1. 01

    Configure once

    goloop configure picks your provider, models, and signs you in — saved to ~/.goloop/config.yaml.

  2. 02

    Set a goal

    goloop init (or run auto-init) captures the objective into the project's .goloop/config.yaml.

  3. 03

    Let it loop

    goloop run . plans, delegates, evaluates, and checkpoints until the objective is met.

// supervisor action — every iteration

{
  "action": "delegate",        // evaluate · ask_user ·
                              // delegate_tools · complete
  "phase":  "build",
  "delegate_task": "Implement add/list/done commands",
  "checkpoint_update": {
    "completed":  ["CLI scaffold"],
    "next_steps": ["persist todos to disk"]
  },
  "status": "success"
}

// 03 — get started

Running in under a minute.

install

# macOS / Linux / WSL
curl -fsSL https://raw.githubusercontent.com/\
mantyx-io/goloop/main/scripts/install.sh | bash

# …or with Go
go install github.com/mantyx-io/goloop/cmd/goloop@latest

configure & run

# once: provider, models, sign-in
goloop configure
goloop login

# in your project
cd your-project
goloop run . --goal "Build a todo CLI"

// supervisors

ChatGPT subscription OpenAI API Anthropic API

// workers

Cursor CLI Claude Code

// 04 — configuration

Plain YAML. Three layers.

global ~/.goloop/config.yaml Provider, models, worker, loop defaults — set once.

project <dir>/.goloop/config.yaml The objective for this repo. Required to run.

run goloop run --goal … One-off overrides for a single invocation.

Give your terminal a goal.

Free and open source under the MIT license.

curl -fsSL …/install.sh | bash