riffpad
Control AI coding agents (Claude Code, Codex, Kimi) from your phone via Riffpad. Use when the user wants to watch, approve, or steer a coding agent remotely / set up Riffpad / pair their phone / keep a long-running agent session going while away from the computer.
供给资产档案
编程与开发 Agent
代码审查、仓库分析、测试、CI、GitHub、DevOps 与开发工作流 Skill。
场景
编程 Agent
我需要一个能理解仓库、修改代码并审查 Pull Request 的编程 Agent。
适配 Agent
Claude Code + OpenAI Agents + Browser agents
适用于 Codex、Claude Code、Cursor、CLI 或自定义 Agent。
安装
就绪
npx skills add riffpad/riffpad --skill riffpad
维护状态
新鲜
今天有推送
风险
高风险
Dependency or permission surface needs review
GitHub 质量
126
68/100 质量 · 73/100 信任
覆盖标签
审查说明
Dependency or permission surface needs review · Permission surface may require sandboxing
Agent 采用评分卡
一眼查看信任、审计与安装准备度
这些分数综合公开仓库元数据、OpenAgentSkill 审查信号、维护新鲜度与安装准备度。它用于候选筛选,不替代人工审查。
质量
有潜力有用的候选项,但采用前应与替代方案比较。
信任
仅限沙盒有用但信任信号不足或混杂的候选项。在结果闭环证明任务匹配前,请保持在隔离工作区内使用。
审计
高风险对安装准备度、安全元数据、维护情况与采用风险的机器可读审查。
OpenAgentSkill 信任评分 v5
仅限沙盒
仅在沙盒中运行,并在用于真实工作前比较接近的替代方案。
Stars
126 个 GitHub Stars
仓库活跃度
126 个 Star,31 个 Fork
维护状态
今天有推送
许可证
Apache-2.0
安装
npx skills add riffpad/riffpad --skill riffpad
安装安全性
标准软件包或运行时安装路径
权限范围
secrets or environment access, shell or command execution
Agent 结果
暂未有 Agent 结果数据
文档
README/SKILL.md 上下文充分
风险摘要
生产前审查
- Financial research output is not financial advice; require human review before any live investment decision.
- This skill may touch real-money trading, broker, wallet, or exchange operations; use only in a sandbox with explicit approval.
- Quality score needs review
- Permission surface needs review: secrets or environment access, shell or command execution
安装准备度
安装路径可用
- 安装路径可用
- 仓库证据可用
- 已声明许可证
- 暂无 Agent 验证结果证据
Agent 可读元数据
这个 Skill 的机器可读决策数据。
使用此区块或内嵌 JSON 判断 Agent 是否应安装该 Skill、选择替代方案,或先请求人工审查。
适用任务
- Local desktop 工作流
- Claude Code 团队
- builders willing to evaluate younger projects
- Navigate local resources
适用 Agent
安装决策
- 命令
- npx skills add riffpad/riffpad --skill riffpad
- 策略
- 阻止
- 人工审查
- 是
信任与风险
- 信任
- 65/100
- 审计
- 78/100
- 风险级别
- 高风险
结果闭环
- 端点
- /api/agent/outcome
- 事件 ID
- resolve
- 结果
- 5
不适用场景
- 需要厂商支持 SLA 的团队
- 没有内部安全审查的高合规环境
- 暂未有 OpenAgentSkill 使用反馈数据
- Audit risk risky exceeds max_risk=medium
- 高风险权限提示:Shell or command execution, Secrets or environment access
替代 Skill
Opencode
200.7K Stars
npx skills add anomalyco/opencode
替代 Skill
Code Review
168.6K Stars
npx skills add mattpocock/skills --skill code-review
替代 Skill
Grill With Docs
164.7K Stars
npx skills add mattpocock/skills --skill grill-with-docs
替代 Skill
To Spec
164.7K Stars
npx skills add mattpocock/skills --skill to-spec
Agent 安全 v2
34/100 · 避免自动安装
This skill should not be selected by an agent without explicit human security review.
Do not auto-install. Inspect the source, dependencies, and permission surface first.
高
Shell 或命令执行
Skill 元数据引用了终端、CLI、Shell、子进程或命令执行工作流。
中
Browser automation
Skill may drive a browser or interact with web pages.
中
网络访问
Skill 可能访问远程页面、API、仓库或外部服务。
中
文件系统访问
Skill 可能读取或写入项目文件、文档、生成产物或本地工作区状态。
- Audit risk risky exceeds max_risk=medium
- 高风险权限提示:Shell or command execution, Secrets or environment access
- Dependency or permission surface needs review
安装目标
在你的 Agent 工作流中安装此 Skill
通过公开安装端点获取命令、安全清单、目标提示词和该 Skill 的规范链接。
OpenAgentSkill CLI
Resolve policy, run the source installer safely, and report a verified install receipt.
$ npx --yes https://github.com/Leon-Drq/openagentskill/releases/download/cli-v0.2.1/openagentskill-0.2.1.tgz install riffpad-riffpadAgent 解析计划
让 Agent 在安装前验证匹配度。
Resolve API 返回首选 Skill、替代方案、安全策略、审计说明、安装目标和可直接执行的提示词,无需抓取此页面。
打开 JSON
/api/agent/resolve?task=Use%20riffpad%20for%20an%20agent%20workflow&agent=codex&max_risk=medium
Resolve 文本
/api/agent/resolve?task=Use%20riffpad%20for%20an%20agent%20workflow&agent=codex&max_risk=medium&format=text
安装交接
/api/skills/riffpad-riffpad/install
Agent 应检查
- 从 Resolve API 检查任务匹配与替代方案。
- 检查审计评分、信任评分和安全策略警告。
- 检查 Codex、Claude Code、Cursor 或 CLI 的安装目标兼容性。
复制提示词
Task: Use riffpad in this workspace.
Resolve first: https://www.openagentskill.com/api/agent/resolve?task=Use%20riffpad%20for%20an%20agent%20workflow&agent=codex&max_risk=medium
Review install handoff: https://www.openagentskill.com/api/skills/riffpad-riffpad/install
Install command: npx skills add riffpad/riffpad --skill riffpad
Before running it, summarize audit warnings, required permissions, and the fallback skill if install is risky.Agent 交接
把安装路径交给 Agent,而不是再给一个目录页。
通过公开安装端点获取命令、安全清单、目标提示词和该 Skill 的规范链接。
安装交接
/api/skills/riffpad-riffpad/install
LLM 文本格式
/api/skills/riffpad-riffpad/install?format=text
寻找替代方案
/api/skills/search?q=riffpad&limit=3
Agent 提示词
Use riffpad for this task. Review https://www.openagentskill.com/api/skills/riffpad-riffpad/install, then install with: npx skills add riffpad/riffpad --skill riffpadRegistry 元数据
用于自动选择 Skill 的 Agent 可读档案。
本页通过 Registry API 提供相同的决策、信任、审计、场景和安装信号,让 Agent 无需抓取界面即可排序。
Manifest
/api/registry/manifest/riffpad-riffpad
LLM 文本
/api/registry/manifest/riffpad-riffpad?format=text
安装别名
/api/registry/install/riffpad-riffpad
推荐
/api/registry/recommend?task=Use%20riffpad%20in%20an%20agent%20workflow&limit=3
适配 Agent
Local desktop
平台
Claude Code, OpenAI Agents, Browser agents
Agent 决策面板
Fallback candidate for Local desktop
先用此 Skill 做原型验证,并保留备选方案。
栈中角色
备选候选
主要匹配
Local desktop
信任标签
先做原型验证
安装路径
命令已就绪
适用场景
- Local desktop 工作流
- Claude Code 团队
- builders willing to evaluate younger projects
证据
- 仓库近期活跃
- 已提供安装命令或 GitHub 仓库
- 68/100 质量档案
先审查
- 暂未有 OpenAgentSkill 使用反馈数据
实施路径
- 1在沙盒 Agent 中安装它,并端到端完成一次Local desktop任务。
- 2Compare output quality, latency, and failure behavior against at least one alternative.
- 3Promote it into production only after reviewing repository permissions, license, and maintenance signals.
信任档案
仅限沙盒
有用但信任信号不足或混杂的候选项。在结果闭环证明任务匹配前,请保持在隔离工作区内使用。
GitHub 采用度
信息126 个 GitHub Stars
Star/Fork 活跃度
检查126 个 Star,31 个 Fork; 当前元数据中没有议题活跃度信息
近期维护
通过今天有推送
许可证清晰度
通过Apache-2.0
积极信号
- AI 审查已通过
- 安装路径可用
- 仓库证据可用
- 近期维护的仓库
- 安装命令未发现明显高风险模式
- 结果闭环已就绪,但需要首次真实 Agent 运行
安装前审查
- Financial research output is not financial advice; require human review before any live investment decision.
- This skill may touch real-money trading, broker, wallet, or exchange operations; use only in a sandbox with explicit approval.
- Quality score needs review
- Permission surface needs review: secrets or environment access, shell or command execution
- Stars/forks activity: 126 stars, 31 forks; issue activity unavailable in current metadata
- Dependency/runtime risk: command execution surface, credential or environment access
- Permission surface: secrets or environment access, shell or command execution
- 暂未有真实 Agent 结果报告
- 无人值守安装前需要人工审查
建议操作
仅在沙盒中运行,并在用于真实工作前比较接近的替代方案。
质量档案
有潜力 适用于 Agent 工作流的候选
有用的候选项,但采用前应与替代方案比较。
工作流匹配
在这些场景使用此 Skill
Operate local tools
Local desktop
I need my agent to operate local files and desktop apps in a repeatable workflow.
Build and ship code
Coding agents
I need a coding agent that can understand a repository, edit code, and review pull requests.
Operate web apps
Browser automation
I need my agent to control a browser, fill forms, and verify web app workflows.
工作流匹配
加入完整工作流
Operate and verify web apps
Browser QA agent
A workflow for agents that navigate products, fill forms, take screenshots, and verify real user flows across web applications.
Inspect, patch, and verify code
Coding review agent
A workflow for software agents that inspect repositories, review pull requests, generate tests, and turn findings into shippable patches.
Design, build, test, and ship interfaces
Frontend and UI
A practical workflow for agents that turn product briefs or Figma designs into polished frontend code, review the result, test it in a browser, and prepare a safe deployment.
替代方案短名单
安装前对比
可能适合该任务的相近 Skill。
Opencode
The open source coding agent.
Code Review
Review a branch or diff against repository standards and the originating spec in two independent analysis passes.
Grill With Docs
A relentless interview that pressure-tests a plan against the codebase, sharpens domain language, and updates CONTEXT.md and ADRs when decisions become durable.
To Spec
Turn the current conversation and codebase context into a structured implementation spec, then publish it to the configured project issue tracker.
概览
--- name: riffpad description: Control AI coding agents (Claude Code, Codex, Kimi) from your phone via Riffpad. Use when the user wants to watch, approve, or steer a coding agent remotely / set up Riffpad / pair their phone / keep a long-running agent session going while away from the computer. ---
# riffpad
Watch, approve, and steer AI coding agents (Claude Code, Codex, Kimi) from your phone. A local daemon bridges the CLI agents running on the user's own machine to their phone, end-to-end encrypted through a zero-knowledge relay. **Not a cloud IDE** — agents keep running on the user's machine with their own API keys and repos; the phone is just a mirror + remote control.
## Install
If `riffpad` is not installed:
```bash # macOS / Linux curl -fsSL https://riffpad.ai/install.sh | sh
# Windows (PowerShell) irm https://riffpad.ai/install.ps1 | iex ```
Verify with `riffpad version`. If still not found after install, open a new terminal.
---
## Onboarding flow (run these in order)
1. **Install** the daemon (above), then `riffpad version` to confirm. 2. **Sign in on the computer**: `riffpad login` - Opens a browser for GitHub authorization (like `gh auth login`). The **user** completes it in the browser; the daemon registers this machine and restarts automatically. - Non-GitHub / password login: set `RIFFPAD_RELAY_USER` + `RIFFPAD_RELAY_PASSWORD`, then `riffpad login --username`. 3. **Pair the phone**: `riffpad pair` - Prints a 6-char code + QR. **The user** must open https://app.riffpad.ai on their phone, sign in with the same account, and enter the code (or scan the QR). You can't complete this step for them — tell them what to do. 4. **Start a session**: `riffpad run codex` (or `claude` / `kimi`) - The terminal stays visible and interactive. The session mirrors to the phone immediately; the user can watch progress, approve permission prompts, and send new instructions from the phone.
> To capture a Claude session the user **already started themselves** (instead of launching a new one), use `riffpad attach` (injects Claude hooks) — not `run`.
---
## CLI reference
| Command | Description | |---|---| | `riffpad login [--url wss://…]` | GitHub device sign-in (default); `--username` for password login | | `riffpad logout` | Sign out and revoke the relay token | | `riffpad auth` | Show the current account (live validation) | | `riffpad pair` | Print a 6-char pairing code and QR | | `riffpad run [claude\|codex\|kimi] [--name N] [--prompt P] [--cwd D]` | Create a hosted session; terminal stays visible/interactive | | `riffpad attach` / `detach` | Inject/remove Claude hooks to capture the user's own running session | | `riffpad sessions` | List sessions | | `riffpad status` | Daemon status | | `riffpad setup [--remove]` | Install/remove the Linux systemd autostart | | `riffpad kill` | Kill switch — stop all sessions + revoke all paired devices | | `riffpad update` | Replace with the latest version (SHA256 + atomic swap) | | `riffpad logs` | Show daemon logs | | `riffpad version` | Print the version |
### `run` options - `--name N` — label the session - `--prompt P` — start with an initial prompt - `--cwd D` — set the working directory
### Languages Output defaults to English. Use `--lang zh` to switch to Chinese; `--lang en` explicitly selects English.
### Environment variables | Variable | Default | Purpose | |---|---|---| | `RIFFPAD_RELAY_URL` | `wss://api.riffpad.ai` | Relay address (change to self-host) | | `RIFFPAD_RELAY_USER` / `RIFFPAD_RELAY_PASSWORD` | — | Password-login credentials | | `RIFFPAD_URL` | `http://127.0.0.1:8787` | Local daemon address | | `RIFFPAD_DIR` | `~/.config/riffpad` | Data directory |
---
## Self-host the relay (optional) The relay is the only server component, and it's zero-knowledge — it forwards ciphertext and stores nothing. To point Riffpad at your own relay, set `RIFFPAD_RELAY_URL` before `login`. Relay source + self-host notes: https://github.com/riffpad/riffpad (`apps/relay`).
---
## Common pitfalls
| Pitfall | Correct approach | |---|---| | Treating it like a cloud IDE | Agents run on the user's machine with their own keys; the phone only mirrors + sends approvals/instructions. Don't try to move the session to the cloud. | | Skipping pairing | `riffpad pair` must run, and the user must complete pairing on their phone at app.riffpad.ai. Without it, nothing reaches the phone. | | Using `run` for a session the user already started | Use `riffpad attach` (injects Claude hooks into an existing session), not `run` (which starts a new one). | | Approvals not showing on the phone | Confirm `riffpad status` is healthy and the phone is paired; check `riffpad logs`. | | `login` can't open a browser (headless) | Use `--username` with `RIFFPAD_RELAY_USER` / `RIFFPAD_RELAY_PASSWORD`. | | Need to stop everything immediately | `riffpad kill` — stops all sessions + revokes all paired devices (kill switch). |
---
## Notes - **Local-first**: code, repos, and API keys never leave the user's machine. - **End-to-end encrypted** (X25519 + AES-256-GCM); the relay is zero-knowledge. - **Read-only by default** on the phone — every approve/reject/prompt is an explicit tap. - Docs: https://www.riffpad.ai/docs/guide/quickstart · Repo: https://github.com/riffpad/riffpad · App: https://app.riffpad.ai
技术详情
- 版本
- 1.0.0
- 许可证
- Apache-2.0
- 最近更新
- 2026年8月23日
- 发布时间
- 2026年8月23日
决策摘要
备选候选
仓库近期活跃
Agent 验证证据
Agent 验证证据
来自解析、审查、安装和一次小范围运行后的结果报告。
- 成功率
- —
- 近期失败
- —
- 结果
- 0
- 输出质量
- —
- 失败
- 0
- 不相关
- 0
- 安装次数
- 0
- 风险拦截
- 0
- 需要配置
- 0
- 生产环境
- 0
暂时没有 Agent 结果数据。首次 Agent 执行可以通过 /api/agent/outcome 报告成功、需要设置、风险拦截、失败或不相关。
增长闭环
分享工具包
为 riffpad 准备的场景化草稿,可手动发布到 X。
riffpad: Control AI coding agents (Claude Code, Codex, Kimi) from your phone via Riffpad. Use when the... 126 stars https://www.openagentskill.com/skills/riffpad-riffpad?ref=x
可选:带安装命令的回复
Listing + install path for riffpad: https://www.openagentskill.com/skills/riffpad-riffpad?ref=x Install: npx skills add riffpad/riffpad --skill riffpad
收录来源
Registry 收录
此列表来自公开来源,维护者认领获批前不会标记为官方。
- 创作者
- riffpad
- 收录方
- OpenAgentSkill 社区索引
归属链接指向公开仓库或创作者主页。创作者可认领列表以更新所有权信号。
认领此 Skill所有者认领
认领此 Skill 页面
这条 Registry 收录 列表归属于 riffpad,但尚未标记为官方。认领后可增加已验证所有者信号,使后续发布、安装和审计更新更值得信赖。
创作者外链工具包
将证据徽章加入你的 README
在开发者评估仓库的位置展示规范页面、当前信任与审计信号,以及真实的 Agent 验证证据。
[](https://www.openagentskill.com/skills/riffpad-riffpad)
[](https://www.openagentskill.com/skills/riffpad-riffpad)
[](https://www.openagentskill.com/skills/riffpad-riffpad/audit)
[](https://www.openagentskill.com/skills/riffpad-riffpad)作者
riffpad
@riffpad
健康信号
- GitHub Stars
- 126
- 质量评分
- 38/100
- 最近 GitHub 推送
- 2026年8月23日
- 框架提示
- 未知
- OpenAgentSkill 浏览量
- 0
- 复制安装命令
- 0
- 跳转点击
- 0
社区信号
告诉我们这个 Skill 是否对你的 Agent 工作流有帮助。汇总反馈会持续改善排序。
信任与安全
仅限沙盒
- GitHub 采用度126 个 GitHub Stars信息
- Star/Fork 活跃度126 个 Star,31 个 Fork; 当前元数据中没有议题活跃度信息检查
- 近期维护今天有推送通过
- 许可证清晰度Apache-2.0通过
- README/SKILL.md 完整度元数据包含足够的用法与工作流上下文通过
- 依赖与运行时风险command execution surface, credential or environment access检查
相关 Skill
Opencode
The open source coding agent.
200.7K StarsCode Review
Review a branch or diff against repository standards and the originating spec in two independent analysis passes.
168.6K StarsGrill With Docs
A relentless interview that pressure-tests a plan against the codebase, sharpens domain language, and updates CONTEXT.md and ADRs when decisions become durable.
164.7K StarsTo Spec
Turn the current conversation and codebase context into a structured implementation spec, then publish it to the configured project issue tracker.
164.7K Stars