← 返回 Skills 市场
Ralph Loops
作者
qlifebot-coder
· GitHub ↗
· v1.0.2
6270
总下载
1
收藏
0
当前安装
3
版本数
在 OpenClaw 中安装
/install ralph-loops
功能描述
Runs autonomous iterative AI loops for requirements, planning, or building phases using structured prompts and fresh context per iteration.
安全使用建议
Install only in an isolated repository or VM with minimal credentials. Review the templates before running, remove or disable automatic git push/tag behavior unless you explicitly want remote publication, and do not expose the dashboard to a network without authentication.
功能分析
Type: OpenClaw Skill
Name: ralph-loops
Version: 1.0.2
This skill is classified as suspicious due to its reliance on `--dangerously-skip-permissions` for the Claude CLI, which grants the AI agent full system access, bypassing all built-in safeguards. Additionally, the `scripts/ralph-loop.mjs` script includes a `checkCmd` argument that allows arbitrary command execution via `execSync`, creating a direct command injection vector if the agent's arguments can be manipulated. The skill also instructs the agent to perform `git push` operations and use the `browser` tool to open URLs (e.g., `browser action=open targetUrl="http://localhost:PORT"`), which are powerful capabilities that, while intended for development, significantly increase the attack surface for prompt injection or misuse. These risky capabilities are present in `SKILL.md`, `scripts/ralph-loop.mjs`, `templates/loop.sh`, and `templates/PROMPT_build.md`.
能力评估
Purpose & Capability
The stated purpose is autonomous iterative development, and the artifacts match that purpose, but the capability set is high impact: Claude is run with permission checks bypassed, prompts direct repository edits and commits, and templates push to git remotes.
Instruction Scope
The skill discloses the permission bypass later in its safety section, but quick-start and template flows can start broad autonomous loops without a strong per-run approval gate; build prompts also allow large-scale repo search, subagents, spec rewrites, tags, commits, and pushes.
Install Mechanism
Installation is conventional for a Node dashboard and Claude Code dependency, with no hidden install-time execution found, but it recommends a global Claude Code version change and npm dependency installation.
Credentials
Running from the user's workspace/home context plus bypassed Claude permissions is broader than ordinary skill execution; the dashboard also reads local loop state and Clawdbot transcripts and enables permissive CORS without authentication.
Persistence & Privilege
The loop is intentionally long-running and writes state to /tmp and archives to ~/clawd/logs/ralph-archive; it does not install an autostart service, but it can persist autonomous work across many iterations and publish changes via git push.
如何使用
- 确保已安装 OpenClaw（本地或 Docker 部署）
- 在对话框中输入安装命令：
/install ralph-loops - 安装完成后，直接呼叫该 Skill 的名称或使用
/ralph-loops触发 - 根据 Skill 的参数说明提供必要输入，即可获得结构化输出
版本历史
v1.0.2
Added warning: don't block conversation while monitoring loops
v1.0.1
Added credits/attribution
v1.0.0
Initial release: autonomous building with Geoffrey Huntley methodology
元数据
常见问题
Ralph Loops 是什么？
Runs autonomous iterative AI loops for requirements, planning, or building phases using structured prompts and fresh context per iteration. 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件，目前累计下载 6270 次。
如何安装 Ralph Loops？
在 OpenClaw 或 Claude Code 对话框中运行命令「/install ralph-loops」即可一键安装，无需额外配置。
Ralph Loops 是免费的吗？
是的，Ralph Loops 完全免费（开源免费），可自由下载、安装和使用。
Ralph Loops 支持哪些平台？
Ralph Loops 跨平台运行，可在任意部署了 OpenClaw / Claude Code 的环境中使用（cross-platform）。
谁开发了 Ralph Loops？
由 qlifebot-coder（@qlifebot-coder）开发并维护，当前版本 v1.0.2。
推荐 Skills