bama
BAMA: Build Any Mac App. Agent skills to research, clone, design, build, optimize, and test native macOS apps with Swift and SwiftUI.
翻译准备中
尚未检测到更新。GitHub 发布版本后会显示在这里。
个人 GitHub 账号下公开仓库的所有者可以登录并验证所有权,自行管理介绍,无需管理员批准。
登录并生成验证文件,在 30 分钟内将其提交到此仓库的默认分支。其他分支或分叉仓库中的文件无效。然后在下方检查文件,即可无需管理员批准直接管理项目。请保留文件以维持管理权限。
如需登录,将前往登录页面。登录后请回到此处,再次点击所选的验证按钮。
在项目文件夹中运行 /nomorevibe verify,即可验证所有权并自行管理介绍。
一分钟完成验证Agent Skills for plori: create and drive cloud AI agents in persistent environments.
翻译准备中
取自仓库 README
正在加载完整 README。 以下为已保存的摘录。
Instructions for coding agents to use plori: a cloud AI agent
with its own persistent environment. The repository also holds general skills that
the plori team wrote for its own work in Claude Code and Codex.
Paste this into an interactive Claude Code conversation:
Read https://plori.ai/.well-known/agent-skills/plori/SKILL.md and install/connect Plori over MCP.查看完整 README ↗Instructions for coding agents to use plori: a cloud AI agent with its own persistent environment. The repository also holds general skills that the plori team wrote for its own work in Claude Code and Codex.
Paste this into an interactive Claude Code conversation:
Read https://plori.ai/.well-known/agent-skills/plori/SKILL.md and install/connect Plori over MCP.
Claude reads the instructions and configures MCP if needed. When a new server has
not loaded, it asks you to type /reload-plugins in the same conversation. With
pairing, open the short address Claude shows, enter the code, verify your email,
and approve the connection. You can approve from a phone while Claude Code runs on
a remote machine.
No installed skill or plugin is required for this flow. Other clients use their own setup steps in the connection guide.
To keep these instructions available across conversations, install the skill:
npx skills add plori-ai/skills
This command installs every skill in the repository. To install one skill, name it:
npx skills add plori-ai/skills --skill subagent-coordination
| Skill | What it teaches |
|---|---|
plori | Connect to plori over MCP, the @plori/cli command, or REST, authenticate (OAuth 2.1 or API key), create agents, invoke them and read replies, answer human-in-the-loop requests, schedule deferred runs, build and run workflows |
subagent-coordination | Run a multi-part coding task as a coordinator: plan, split the work into units that own separate files, brief up to five workers (native subagents plus tasks in the other local coding CLI, with a wrapper that lets you message a running Codex worker), review each result before you accept it, and ship one PR |
root-cause-debugging | Find the root cause of an incident or a recurring bug: build an evidence timeline, answer three causal questions, check current upstream guidance, and design a fix for the mechanism, not the symptom |
record-live-demo | Record a real terminal, browser, or desktop product demo, then verify it and make an H.264 MP4 for delivery, with trim and speed helpers |
plori skill mirrors the canonical copy served at
https://plori.ai/.well-known/agent-skills/ (Agent Skills Discovery RFC); both are
generated from the same source and updated together on every plori release.subagent-coordination, root-cause-debugging, and record-live-demo are
general skills. They do not need a plori account.https://api.plori.ai/mcp for clients with
native MCP support, and ships a CLI (@plori/cli,
npm i -g @plori/cli) for terminal agents; the skill covers all three paths.MIT