Skill 与 AgentSKILLS & AGENTS· CLAUDE-SKILL
loopx-project/loopx
让AI代理能持续运行数小时甚至数天的控制面板。A control plane that keeps AI coding assistants running for hours, turning long-term goals into verified results.
根据公开文档整理,未运行实测;费用以项目官网为准。Based on public documentation, without runtime testing. Check the project site for pricing.
能力CAPABILITIES
加速项目中最慢的流程
持续运行研究任务
通过仪表盘查看进度
比较方法并验证改进
管理长期目标与协作
Set a goal and let AI work on it for hours unattended
See evidence of each step before accepting changes
Run benchmarks across multiple configurations automatically
Connect AI to local databases and tools like Playwright
Review and approve multi-step deliveries from a dashboard
上手GET STARTED
git clone https://github.com/loopx-project/loopx ~/loopx选择与你的环境对应的方式。命令来自项目文档,未运行实测。Choose the option for your environment. Commands come from project documentation and have not been run-tested.
使用限制LIMITATIONS
以下要求与风险来自文档静态检查,未执行项目代码;扫描没有命中,不代表使用安全。Risk here is evidence, not an alarm. These findings come from static scanning of the docs; we never execute a project’s code.
限制LIMITS
需检查Needs review
静态扫描STATIC SCAN
01控制浏览器Controls a browserbenchmark/swe-marathon/skills/tb4-five-arm/SKILL.md
还要 MCP server(playwright,sse `http://playwright-mcp:3080/sse`)。
扫描发现项目包含控制浏览器的MCP server(Playwright),可能允许自动化浏览器操作,存在潜在安全风险。建议用户在使用时注意权限控制,并自行审计相关代码。The static scan found a medium-risk item: LoopX can control a browser via Playwright. This is expected for tasks that need web interaction, but you should review how it's used in your setup.
项目动态PROJECT ACTIVITY
首次发现FIRST SEEN
14 NOV 202614 NOV 2026
增长GROWTH
+2.2%+2.2%
6.6k → 6.7k stars · 7 个快照6.6k → 6.7k stars · 7 snapshots
状态STATUS
持续活跃Active
质量 AQuality A
LoopX 让你用自然语言给 AI 助手布置一个长期目标(比如“优化项目里最慢的流程”),它会自己拆任务、反复执行、验证结果,你随时可以用 `loopx dashboard` 看进度。Over 6,600 stars and 750+ commits in the last week show rapid adoption and active development.
{
"agents": [
{
"agent": "claude-code",
"evidence": "readme: …lways-on agents across Codex, Claude Code, DeepSeek Harness and other r…"
},
{
"agent": "codex",
"evidence": "readme: …the runtimes you already use (Codex CLI, Claude Code, DeepSeek Harnes…"
},
{
"agent": "opencode",
"evidence": "readme: …after strict verification | | OpenCode | Install the static command…"
}
],
"apiKey": null,
"docker": null,
"taxonomy": {
"scores": {
"design": 0,
"ai-tools": 1,
"dev-tools": 1,
"indie-web": 0,
"skill-agent": 6,
"productivity": 3
},
"primary": "skill-agent",
"secondary": "claude-skill"
},
"localRuntime": "r.md). - **Independent user · vLLM optimization merged upstream.** [Zhen",
"skillMdTotal": 16,
"skillMdErrors": [],
"treeTruncated": false,
"categoryScores": {
"data": 4,
"agent": 10,
"media": 0,
"design": 0,
"browser": 0,
"devtool": 1,
"security": 0,
"marketing": 0,
"productivity": 0
},
"classification": [
{
"signal": "16 SKILL.md with name + description",
"weight": 0.95
},
{
"signal": "skills/ directory with 22 markdown files",
"weight": 0.75
},
{
"signal": "MCP SDK dependency",
"weight": 0.6
},
{
"signal": "description mentions AI",
"weight": 0.2
}
],
"scannedSources": [
"readme",
"apps/presentation/site/public/skill.md",
"apps/workbench/skills/loochat-ask/SKILL.md",
"apps/workbench/skills/loochat-tools/SKILL.md",
"benchmark/swe-marathon/skills/swe-marathon-five-arm/SKILL.md",
"benchmark/swe-marathon/skills/tb4-five-arm/SKILL.md",
"loopx/capabilities/manager_context/skills/loopx-manager/SKILL.md",
"skills/loopx-benchmark/SKILL.md",
"skills/loopx-change-quality/SKILL.md",
"skills/loopx-doc-registry/SKILL.md",
"skills/loopx-material/SKILL.md",
"skills/loopx-performance-diagnosis/SKILL.md",
"skills/loopx-pr-merge/SKILL.md",
"skills/loopx-pr-program/SKILL.md",
"skills/loopx-pr-review/SKILL.md",
"skills/loopx-project/SKILL.md",
"skills/loopx-self-repair/SKILL.md"
],
"skillMdFetched": 16
}