Codeg 多智能体协同工作的平台
Collaborative multi-agent AI coding workspace: aggregate sessions from Claude Code, Codex, OpenCode, Pi, Grok Build, etc. Desktop app, self-hosted server, or Docker.
4.1K
DEV TRENDS / CATEGORY
Programming language · Rust
18 new projects first discovered during Oct 21, 2026 – Oct 27, 2026, deduplicated across 6 sources.
150 projects first discovered Aug 5, 2026 – Oct 27, 2026
DTDevTrends classificationOriginalOriginal project or source tag
18 new projects6 sources
：公众号文章采集工具，https://github.com/areyoubugcoder/mpider
Windows 微信公众号采集与离线阅读工作台 · 原生会话协议、检查点续采、Markdown / HTML 导出
Original
30
开源了一个 MCP 治理网关： Jev 判 0.94、d3 判 0.70，两个判官说不同的话时怎么办
给 AI agent 做了个 MCP 执法网关（ Rust ），核心问题：当两个决策模型对同一件事给出不同判断时，怎么安全处理？ 实测：同一个越权场景（ coding agent 试图 merge 夹带无关变更的 PR ）： - Jev 给 0.94 置信度 → 准入层 Accept → 授权挂起 → DENY - d3 （ vLLM SR ）给 0.70 → Hold → 什么都不改 → ALLOW 两个都对。分歧不是 bug ，是校准本性（ d3 ECE 4.3 vs Jev 8.0 ）。治理层的价值在于让分歧安全。 网关做的事：每个 tools/call 先过 r2r_decide 门控，ALLOW 转发上游，DENY 返回 isError + 证据文本（相位 + 规则 + 事件 ID ），全程审计到 EventStore 。 零 API key 复现： cargo run -- fixture cargo run -- fixture --provider d3 GitHub: （ Apache-2.0, v0.2.0 ）
Original
—
Find overlapping Rust tests with mutation testing
To tackle the ever-increasing number of tests, I built cargo-test-prune. The idea behind it is to use cargo-mutants to generate mutants, and then solve the set cover problem for those mutants to suggest which tests can be eliminated. (use a greedy algorithm to cover the detected mutants with fewer tests) I came up with this as a countermeasure against the monotonically increasing tests that tend to accumulate, especially when coding with AI. In fact, this very repository was also written using AI. I ran this tool on a previous revision of this repository. There were 11 tests and 108 generated mutants: 73 were caught by the tests, 23 survived, and 12 failed to compile. I removed the 5 tests suggested by the tool and re-ran the same 108 mutants. The remaining 6 tests caught the exact same 73 mutants. Tests partly serve as documentation, so simply applying this isn't necessarily recommended, but since writing less code still holds value in the AI era, I think this tool could be quite useful.
0
MacLean — Safe Computer Cleanup for Developers
## maclean —— 给开发者的开源磁盘清理 CLI（Rust） maclean 是一个用 Rust 编写、面向开发者的磁盘清理与存储智能命令行工具。 扫描结果会明确标注「删了会怎样」——要重新下载、要重新编译，还是涉及用户数据需确认； 删除前对受保护应用单独拦一道，缓存永久删除、大文件移入废纸篓可还原。 ### 它解决什么问题 普通清理工具只告诉你「这里占了多少」，却不告诉你「删了后果是什么」。 maclean 把删除决策交给事实：每一项都标注可恢复性与影响面，默认只预览、不擅自删除， 适合脚本、CI/CD、定时任务与自动化工作流。 ### 核心特性 - **扫描 / 分类 / 安全清理**：dev 缓存、大文件、重复项、启动项、已安装应用清单等 - **删除前安全闸门**：三级应用保护、符号链接 / 路径穿越 / 挂载点 / TOCTOU 复检，受保护应用不给删 - **可逆**：大文件进废纸篓，`backups` + `restore <id>` 可还原 - **机器可读**：`--format human|json|jsonl`，稳定退出码，适合管道与 CI - **系统级周期清理**：`schedule` 每 1 / 7 / 30 天自动跑一次安全档清理（launchd / 任务计划，app 不用开） - **双平台**：macOS / Windows，CI 双端构建（纯 CLI，无 GUI、无账号） ### 快速开始 ```bash # Homebrew brew install jamesfeng2009/maclean/maclean # 预览将清理的项目（不删任何文件） maclean clean --tab dev-cache --dry-run # 确认删除（脚本必须显式 --yes） maclean clean --tab dev-cache --yes --format jsonl # 每 7 天自动跑一次安全档清理（系统级，app 无需常驻） maclean schedule --install --days 7 欢迎提交issue
Original
0
Borg
Borg is your personal assistant, employee, developer, analyst, and sidekick. Security-first by default.
3
Enzyme – a programming language that compiles wikis for agents
Local-first compile step for knowledge bases. Save 350x cost, 1000x speed vs. frontier models
96
wolf：基于 llama.cpp 的 Rust 本地推理工具，支持多轮对话与上下文保存
A Rust local GGUF model tool built on llama.cpp. Chat, inspect, and manage context. Agent-ready backend.
1
sslx
A fast, modern alternative to OpenSSL's CLI - inspect, grade, and manage certificates from the terminal
99
Tessera 可自托管的文件分享服务
Self-hosted file and note sharing that cleans up after itself.
62
XTrace – productivity observability tool for coding agents
Open-source macOS app for coding agent productivity: agent vs estimated human time, API-equivalent spend, usage limits, and customizable scoring skills.
25
Rim：一个非稳定类 git 版本管理工具
想要用 AI 提效，却面临自己写好的设计被 AI 改烂、或 slop code 塞满 commit history 的担忧？ - 只是想测试不同的可能性，却不得不创建新 branch 挤满 list ？ - 想要用 agent 简单看看不同方向做出来什么样，但每个方向都要单独建一个 workspace ，最后管理与记住文件夹花费的精力比真正审核代码还多？ 那么，rim 或许适合你。 它是介于 git 与 agent 之间的轻量中间层基础设施，基于 AI 时代「非稳定代码比例持续增加」的理念，不将每一个文件修改都当成高价值、需要完整追溯的 commitment ，而以节点、分叉为版本追溯核心。 使用 rim ，你可以： - 让几个 AI 同时生成不同可能方案，在完成后自己或用 AI 对比验收 - 测试多个方向，选定某个方向后一键剪枝，不留垃圾 - 让 git commit 真正回归只追溯重要与稳定更改，而非留下大量低价值 history - 让 git branch 记录重要的任务分支，而非不断来回切换、塞满每一个细枝末节的并行方向测试 它就在你与下一个 commit 之间。 Rim 正在 alpha 版本测试中，欢迎试用和 feedback （需求、体验和项目方向都欢迎）。
Original—
Knovum：本地优先的无限画布 AI 知识空间（Tauri 2 + Vue 3）
The local-first AI knowledge workspace — an infinite canvas for research, learning, and creation (Tauri 2 + Vue 3)
11
Cathode, a Gtk4 YouTube player based on mpv and yt-dlp
I've wanted this for years and didn't understand why no one had made one: a YouTube client that I can actually use daily. It has a queue of videos, it has local subscriptions, it downloads all the videos fully before playing, so no buffering ever (and downloads everything that is the queue), it has a restricted mode that works great for children, a bunch of other niceties for sane consumption of YouTube content in our age, written in Rust.
1
Voltip：识别可以完全在本机的跨平台语音输入工具
Push-to-talk voice typing for Windows, macOS, Linux and Android. On-device or cloud speech recognition, AI polish with presets, a dictionary, rules and scenes, long recordings with subtitles; the phone works on its own or pairs end-to-end encrypted as the computer's microphone and keyboard.
18
den：终极文档维护方案
我被项目文档维护恶心到极点了，于是返回第一性原理思考文档的本质。它是项目的长期事实记录，需要能被很快地消化吸收。 现有方法论多数是规范化文档治理方法，通过统一心智模型来加速文档的吸收速度。但并没有一个适配所有场景的标准方法论，即使是大厂也经常出现文档过时或内部有多套方法论的情况。 从输入端不好解决，输出端能否解决呢？我马上想到如果可以通过语义搜索项目文档里我需要的内容，那完全没必要死磕文档维护方法了，全部平铺在 docs/下都可以，需要查文档的时候搜索就好嘛。 den 的主要场景就是这个：快速搜索项目内文档。搞完这个心情也舒畅很多
Original—
Twine is a meta-harness with workflows, cross-agent memories and traces
Agentic Development Environment for macOS
36
Dayside：用实时天色看对方醒没醒的 Mac 菜单栏世界时钟
Before you message someone in Tokyo, look at their sky.
3
No matching projects
Try another keyword or category.