能力标签
物联网AI编码
⚙️
Agent工作流

物联网AI编码

基于 TypeScript · 无代码搭建完整 AI 自动化流程
英文名:Adsum-IoT-Coder
⭐ 13 Stars 💻 TypeScript 📄 Apache-2.0 🏷 AI 8.0分
8.0AI 综合评分
ai-assistantiottypescript
✦ AI Skill Hub 推荐

AI Skill Hub 强烈推荐:物联网AI编码 是一款优质的Agent工作流。AI 综合评分 8.0 分,在同类工具中表现稳健。如果你正在寻找可靠的Agent工作流解决方案,这是一个值得深入了解的选择。

📚 深度解析

物联网AI编码 是一套完整的 AI Agent 自动化工作流方案。随着 AI 能力的不断提升,基于 Agent 的自动化工作流正在成为提升个人和团队效率的核心方式。区别于传统的 RPA 自动化(模拟鼠标键盘操作),AI Agent 工作流通过理解任务意图、动态规划执行路径,能够处理更复杂的非结构化任务。

物联网AI编码 工作流的设计遵循"最小配置,最大复用"原则:核心逻辑已经封装好,用户只需配置自己的 API Key 和业务参数即可快速上手。工作流内置错误处理和重试机制,在网络波动或 API 限速等情况下仍能稳定运行,适合作为生产环境的自动化基础设施。

在实际部署时,建议先在测试环境中运行 3-5 次,验证各个环节的输出结果符合预期,再部署到生产环境。AI Skill Hub 评分 8.0 分,是同类 Agent 工作流中的精选推荐。

📋 工具概览

物联网AI编码 是一套完整的 AI Agent 自动化工作流方案。通过可视化的节点编排,将复杂的多步骤任务拆解为清晰的自动化流程,实现全程无人值守的智能处理。支持与数百种外部服务和 API 无缝集成,适合构建数据处理管线、业务自动化和 AI 辅助决策系统。

GitHub Stars
⭐ 13
开发语言
TypeScript
支持平台
Windows / macOS / Linux
维护状态
轻量级项目,按需更新
开源协议
Apache-2.0
AI 综合评分
8.0 分
工具类型
Agent工作流
Forks

📖 中文文档

以下内容由 AI Skill Hub 根据项目信息自动整理,如需查看完整原始文档请访问底部「原始来源」。

物联网AI编码 是一套完整的 AI Agent 自动化工作流方案。通过可视化的节点编排,将复杂的多步骤任务拆解为清晰的自动化流程,实现全程无人值守的智能处理。支持与数百种外部服务和 API 无缝集成,适合构建数据处理管线、业务自动化和 AI 辅助决策系统。

📌 核心特色
  • 可视化 Agent 工作流编排,无需编写复杂代码
  • 支持多步骤自动化任务链,实现全流程无人值守
  • 与外部 API、数据库和第三方服务无缝集成
  • 内置错误处理与自动重试机制,保障稳定运行
  • 提供可复用的自动化模板,快速在同类场景部署
🎯 主要使用场景
  • 自动化日常重复性工作,将精力集中于创造性任务
  • 构建数据采集 → 处理 → 输出的完整自动化管线
  • 实现跨平台、跨系统的数据流转和业务协同
以下安装命令基于项目开发语言和类型自动生成,实际以官方 README 为准。
安装命令
# 方式一:npm 全局安装
npm install -g adsum-iot-coder

# 方式二:npx 直接运行(无需安装)
npx adsum-iot-coder --help

# 方式三:项目依赖安装
npm install adsum-iot-coder

# 方式四:从源码运行
git clone https://github.com/adsumnetworks/Adsum-IoT-Coder
cd Adsum-IoT-Coder
npm install
npm start
📋 安装步骤说明
  1. 访问 GitHub 仓库获取工作流文件
  2. 在对应平台(Dify / Flowise / Make 等)中找到「导入工作流」功能
  3. 上传工作流文件
  4. 按照提示配置必要的环境变量和 API Key
  5. 运行测试确认流程正常后投入使用
以下用法示例由 AI Skill Hub 整理,涵盖最常见的使用场景。
常用命令 / 代码示例
# 命令行使用
adsum-iot-coder --help

# 基本用法
adsum-iot-coder [options] <input>

# Node.js 代码中使用
const adsum_iot_coder = require('adsum-iot-coder');

const result = await adsum_iot_coder.run(options);
console.log(result);
以下配置示例基于典型使用场景生成,具体参数请参照官方文档调整。
配置示例
# adsum-iot-coder 配置说明
# 查看配置选项
adsum-iot-coder --config-example > config.yml

# 常见配置项
# output_dir: ./output
# log_level: info
# workers: 4

# 环境变量(覆盖配置文件)
export ADSUM_IOT_CODER_CONFIG="/path/to/config.yml"
📑 README 深度解析 真实文档 完整度 57/100 查看 GitHub 原文 →
以下内容由系统直接从 GitHub README 解析整理,保留代码块、表格与列表结构。

简介

<img src="assets/icons/icon.png" width="110" alt="Adsum IoT Coder" />

About

Adsum Networks are embedded engineers who shipped IoT gateways and devices on Nordic nRF and other SoC platforms for nine years before shipping an agent. General coding agents leave embedded developers without reliable help for the two jobs that fill the day: the routine setup worth automating, and the runtime bugs that never show up in source review. Adsum does not replace the embedded engineer; it accelerates them.

Why it exists → · How it works → · The team →

What's New <sup>`v0.4.0`</sup>

A home that reads your desk, cellular and satellite work behind a free account, and a run you can steer while it works. BLE, Wi-Fi and Ethernet need no account, today or tomorrow.

<img src="assets/icons/whatsnew-detect.png" width="18" valign="middle" alt="" /> &nbsp;A home that reads your desk. One row says what is on it and what is missing (nRF ✓ nRF52840 DK, or no toolchain yet · what to install →); a device that will not answer is reported there, with why →. Then the runs worth starting, ranked, each saying why. Typing in the box is the new session; a named resume is offered whenever the folder has one, and every other session is one click away in the editor's History. Getting started

<img src="assets/icons/whatsnew-esp.png" width="18" valign="middle" alt="" /> &nbsp;Gateway firmware you can flash and license. Signed images and licensed source for the Fanstel LEW840x on BLE, Ethernet and Wi-Fi; a cellular demo on LTE-M / NB-IoT; satellite NB-NTN and nRF54 edge AI with a free account (GitHub or email, no card). The Fanstel BLG20, a BLE 6 terrestrial + non-terrestrial gateway with on-device inference, is next. Gateway firmware

<img src="assets/icons/whatsnew-byok.png" width="18" valign="middle" alt="" /> &nbsp;Steer a run without stopping it. A message sent to a working session lands at its next step. The thinking depth you set for DeepSeek and GLM now reaches the request, DeepSeek prices match the vendor's, and modelPricing takes your own rate. Models

</td> <td width="42%" valign="top"><img src="assets/docs/home-0.4.0.png" width="100%" alt="The Adsum IoT Coder home in VS Code: the detected nRF52840 DK, the Fanstel LEW840x project, ranked suggested runs, and the cellular and gateway runs that unlock with a free account" /></td> </tr> </table>

In v0.3.1: nRF91 cellular, the first partner open-hardware gateway, and Tool bits downloaded on demand. In v0.2.0: a two-chip industrial gateway built, debugged and CRA-checked from one spec in under 30 minutes. Watch the playlist · full history in the changelog.

Getting Started

Search Adsum IoT Coder in the VS Code Extensions panel, or install from the VS Code Marketplace or Open VSX (Cursor, Windsurf, VSCodium), on Windows, macOS and Linux. No key, no account, no card: the free tier is on by default and is a real working tier, enough to scaffold a project and run a full debug loop.

Prerequisites: the nRF Connect Extension Pack for nRF, or an ESP-IDF installation for ESP. nRF54LM20 needs nRF Connect SDK 3.3.0 or newer; cellular needs an nRF91 DK and a SIM. Full requirements

  1. Start with a sample run, no board needed: a real BLE bug, the same bug one layer deeper with a sniffer and HCI tracing, or a CRA readiness check on a pre-built gateway.
  2. Open your nRF or ESP project. The home detects your boards and toolchains and ranks the runs worth starting: Build, flash & debug, Add a feature, Test & validate, CRA SBOM & Fix, or a guided partner-gateway build.
  3. Describe the task or pick a run. Enter starts the session; a message sent while it works lands at the next step.
  4. Register (free, no card) when you reach cellular, satellite or edge AI. Everything else keeps working without it.
  5. Bring your own model whenever you want: the GLM Coding Plan, Claude, DeepSeek, or any OpenAI- or Anthropic-compatible endpoint, cloud or local (Ollama, LM Studio), switched instantly on a running task.

Field-tested on our own builds: the budget tiers handle routine work with thinking on; the full models can switch it off, which is where the token saving lives. Free tier → · Models →

What it does: debug, build, and prototype ESP and nRF firmware

  • Tells you what is on your desk. Names your board, says which toolchain is missing and what to install, and when a serial device will not answer it says so in the header, with the why one click away.
  • Carries a full product build. One spec to a working two-chip gateway, across both toolchains, in under 30 minutes, you approving each step. Walkthrough
  • Builds, flashes and debugs on real hardware. Live RTT (J-Link) and UART logs on nRF, serial on ESP, read against your source. The reset vector is checked before an image is flashed.
  • Debugs across three layers. App log, HCI bus and over-the-air radio (an nRF52840 dongle the agent flashes as an nRF Sniffer for Bluetooth LE), correlated. A real one: 36 advertisements on the air, 0 received, a radio front end never switched on.
  • Talks to the board. AT and Zephyr shell commands, and a modem trace with the network's own reason for refusing a connection.
  • Scaffolds, extends, tests. A new project, or a BLE service, sensor, shell or storage wired into yours; host tests and on-hardware checks.
  • Remembers and can be steered. An .adsum/ project memory read at the start of every task; a message mid-run lands at the next step; any session exports as one redacted file.

Supported platforms: ESP32 / ESP-IDF and nRF / nRF Connect SDK

PlatformChips (today)SDKProtocols (today)
**Nordic**nRF52, nRF53, nRF54L (L15, LM20), **nRF91 (9160, 9161, 9151)**nRF Connect SDK (Zephyr)BLE, **NB-IoT, LTE-M, GNSS, NB-NTN**
**Espressif**ESP32, ESP32-S3, ESP32-C6, and the rest of the shipping rangeESP-IDFWi-Fi, BLE
**Products****Fanstel LEW840X, BWG840X** gateways, with their own product knowledgeboth, one workspaceBLE, Ethernet, Wi-Fi, cellular
**Roadmap**nRF7x (Wi-Fi), on-device AI on nRF54 / ESP32, Linux devices (NVIDIA Jetson, Raspberry Pi)LoRa, 5G / 5G RedCap via hats

Adsum works on any board built with a supported chip: your own design, a reference board, a development kit, or a product off the shelf. There is no list your board has to be on. Cellular, NB-NTN and edge-AI runs, and the gateway firmware below, need a free registered account; nothing else does. NTN firmware is a public download but runs only on an nRF9151 of the LACA A1A revision; DECT NR+ knowledge ships and needs a modem image from Nordic sales. Chips and protocols · cellular · partner open hardware

Do I need an API key or an account?

No. The free tier is on by default, with no key and no card. A free account (GitHub or email) is needed only for cellular, satellite NB-NTN, nRF54 edge AI and the gateway firmware images.

Which SDK versions are verified?

The gateway builds are verified with nRF Connect SDK 3.2.1 and ESP-IDF 5.5 on extension 0.4.0; nRF54LM20 needs nRF Connect SDK 3.3.0 or newer. Getting started

Does it replace the nRF Connect for VS Code or ESP-IDF extensions?

No. It works beside them: it needs the nRF Connect Extension Pack for nRF and an ESP-IDF installation for ESP, and drives west, nrfutil, idf.py and esptool through those toolchains.

FAQ: editors, boards, accounts, SDK versions

🎯 aiskill88 AI 点评 A 级 2026-05-26

高效的物联网AI编码工具,值得关注

📚 实用指南(长尾问题)
适合谁
  • 构建多智能体协作系统的 Agent 开发者
最佳实践
  • 本地部署优先选 GGUF 量化模型,节省显存并保持响应速度
  • Agent 任务先做 dry-run 验证工具调用链,再开启自主执行
常见错误
  • API key 直接提交到 git 仓库(请用 .env 并加入 .gitignore)
  • 显存不足直接 OOM — 优先降低 context 或换更小的量化模型
部署方案
  • CLI:直接 npm install -g / pip install,命令行调用
  • 本地部署:CPU 8GB 起,GPU 推荐 16GB+ 显存
  • 云端托管:可放在 Vercel / Railway / Fly.io 等 PaaS 平台

⚡ 核心功能

  • 可视化 Agent 工作流编排,无需编写复杂代码
  • 支持多步骤自动化任务链,实现全流程无人值守
  • 与外部 API、数据库和第三方服务无缝集成
  • 内置错误处理与自动重试机制,保障稳定运行
  • 提供可复用的自动化模板,快速在同类场景部署
👥 适合谁
  • 构建多智能体协作系统的 Agent 开发者
⭐ 最佳实践
  • 本地部署优先选 GGUF 量化模型,节省显存并保持响应速度
  • Agent 任务先做 dry-run 验证工具调用链,再开启自主执行
⚠️ 常见错误
  • API key 直接提交到 git 仓库(请用 .env 并加入 .gitignore)
  • 显存不足直接 OOM — 优先降低 context 或换更小的量化模型

👥 适合人群

自动化工程师和运维人员项目经理和业务分析师希望减少重复性工作的专业人士数字化转型团队

🎯 使用场景

  • 自动化日常重复性工作,将精力集中于创造性任务
  • 构建数据采集 → 处理 → 输出的完整自动化管线
  • 实现跨平台、跨系统的数据流转和业务协同

⚖️ 优点与不足

✅ 优点
  • +Apache-2.0 协议,可免费商用
  • +大幅减少重复性人工操作
  • +可视化流程,清晰直观
  • +可扩展性强,支持复杂场景
⚠️ 不足
  • 初始配置和调试需投入一定时间
  • 强依赖外部服务的稳定性
  • 复杂场景需具备一定技术基础
⚠️ 使用须知

AI Skill Hub 为第三方内容聚合平台,本页面信息基于公开数据整理,不对工具功能和质量作任何法律背书。

建议在沙箱或测试环境中充分验证后,再部署至生产环境,并做好必要的安全评估。

📄 License 说明

✅ Apache 2.0 — 宽松开源协议,可商用,需保留版权声明和 NOTICE 文件,含专利授权条款。

❓ 常见问题 FAQ

Adsum-IoT-Coder 是一款TypeScript开发的AI辅助工具。开源AI工作流:Adsum IoT Coder is the AI coding agent purpose-built for IoT firmware — it captu。⭐13 · TypeScript 主要应用场景包括:物联网设备固件开发。
💡 AI Skill Hub 点评

总体来看,物联网AI编码 是一款质量优秀的Agent工作流,在同类工具中具备一定竞争力。AI Skill Hub 将持续追踪其更新动态,建议收藏备用,结合自身场景选择合适时机引入使用。

⬇️ 获取与下载
⬇ 下载源码 ZIP

✅ Apache-2.0 协议 · 可免费商用 · 直接从 aiskill88 服务器下载,无需跳转 GitHub

📚 深入学习 物联网AI编码
查看分步骤安装教程和完整使用指南,快速上手这款工具
🌐 原始信息
原始名称 Adsum-IoT-Coder
原始描述 开源AI工作流:Adsum IoT Coder is the AI coding agent purpose-built for IoT firmware — it captu。⭐13 · TypeScript
Topics ai-assistantiottypescript
GitHub https://github.com/adsumnetworks/Adsum-IoT-Coder
License Apache-2.0
语言 TypeScript
🔗 原始来源
🐙 GitHub 仓库  https://github.com/adsumnetworks/Adsum-IoT-Coder

收录时间:2026-05-26 · 更新时间:2026-05-30 · License:Apache-2.0 · AI Skill Hub 不对第三方内容的准确性作法律背书。