loop by alirezarezvani/claude-skills
npx skills add https://github.com/alirezarezvani/claude-skills --skill loop启动一个按用户选择的时间间隔运行的循环实验。
/ar:loop engineering/api-speed # 启动循环(会提示选择间隔)
/ar:loop engineering/api-speed 10m # 每 10 分钟运行一次
/ar:loop engineering/api-speed 1h # 每小时运行一次
/ar:loop engineering/api-speed daily # 每天约上午 9 点运行
/ar:loop engineering/api-speed weekly # 每周一约上午 9 点运行
/ar:loop engineering/api-speed monthly # 每月 1 号约上午 9 点运行
/ar:loop stop engineering/api-speed # 停止一个正在运行的循环
如果未指定实验,则列出实验列表供用户选择。
如果未提供间隔参数,则显示选项:
选择循环间隔:
1. 每 10 分钟 (快速 — 实时观察)
2. 每小时 (后台 — 稍后查看)
3. 每天约上午 9 点 (过夜实验)
4. 每周一 (长期实验)
5. 每月 1 号 (慢速实验)
映射到 cron 表达式:
| 间隔 | Cron 表达式 | 简写 |
|---|---|---|
| 10 分钟 |
广告位招租
在这里展示您的产品或服务
触达数万 AI 开发者,精准高效
*/10 * * * *10m |
| 1 小时 | 7 * * * * | 1h |
| 每日 | 57 8 * * * | daily |
| 每周 | 57 8 * * 1 | weekly |
| 每月 | 57 8 1 * * | monthly |
使用 CronCreate 并填入以下提示(填写实验详情):
您正在运行自主研究实验 "{domain}/{name}"。
1. 读取 .autoresearch/{domain}/{name}/config.cfg 获取:target, evaluate_cmd, metric, metric_direction
2. 读取 .autoresearch/{domain}/{name}/program.md 获取策略和约束
3. 读取 .autoresearch/{domain}/{name}/results.tsv 获取实验历史
4. 运行:git checkout autoresearch/{domain}/{name}
然后执行恰好一次迭代:
- 查看 results.tsv:哪些成功、哪些失败、哪些尚未尝试
- 对目标文件进行一项修改(根据运行次数进行策略升级)
- 提交:git add {target} && git commit -m "experiment: {description}"
- 评估:python {skill_path}/scripts/run_experiment.py --experiment {domain}/{name} --single
- 读取输出(KEEP/DISCARD/CRASH)
规则:
- 每次实验只做一项修改
- 绝不修改评估器
- 如果 results.tsv 中连续出现 5 次崩溃,则删除此定时任务(CronDelete)并发出警报
- 每完成 10 次实验后,更新 program.md 的策略部分
当前最佳指标:{从 results.tsv 读取或 "尚无基线"}
迄今为止总实验次数:{根据 results.tsv 统计}
写入 .autoresearch/{domain}/{name}/loop.json:
{
"cron_id": "{来自 CronCreate 的 id}",
"interval": "{用户选择}",
"started": "{ISO 时间戳}",
"experiment": "{domain}/{name}"
}
已为 {domain}/{name} 启动循环
间隔:{间隔描述}
定时任务 ID:{id}
自动过期时间:3 天(CronCreate 限制)
查看进度:/ar:status
停止循环:/ar:loop stop {domain}/{name}
注意:定时任务在 3 天后自动过期。
过期后重新运行 /ar:loop 以重启。
当用户运行 /ar:loop stop {experiment} 时:
.autoresearch/{domain}/{name}/loop.json 获取定时任务 IDCronDeleteloop.json/ar:loop 来重启。结果会保留——新的循环会从旧循环停止的地方继续。autoresearch/{domain}/{name} 分支)。每周安装次数
159
代码仓库
GitHub 星标数
6.5K
首次出现
11 天前
安全审计
安装于
codex154
gemini-cli153
kimi-cli152
amp152
github-copilot152
opencode152
Start a recurring experiment loop that runs at a user-selected interval.
/ar:loop engineering/api-speed # Start loop (prompts for interval)
/ar:loop engineering/api-speed 10m # Every 10 minutes
/ar:loop engineering/api-speed 1h # Every hour
/ar:loop engineering/api-speed daily # Daily at ~9am
/ar:loop engineering/api-speed weekly # Weekly on Monday ~9am
/ar:loop engineering/api-speed monthly # Monthly on 1st ~9am
/ar:loop stop engineering/api-speed # Stop an active loop
If no experiment specified, list experiments and let user pick.
If interval not provided as argument, present options:
Select loop interval:
1. Every 10 minutes (rapid — stay and watch)
2. Every hour (background — check back later)
3. Daily at ~9am (overnight experiments)
4. Weekly on Monday (long-running experiments)
5. Monthly on 1st (slow experiments)
Map to cron expressions:
| Interval | Cron Expression | Shorthand |
|---|---|---|
| 10 minutes | */10 * * * * | 10m |
| 1 hour | 7 * * * * | 1h |
| Daily | 57 8 * * * | daily |
| Weekly | 57 8 * * 1 | weekly |
| Monthly |
Use CronCreate with this prompt (fill in the experiment details):
You are running autoresearch experiment "{domain}/{name}".
1. Read .autoresearch/{domain}/{name}/config.cfg for: target, evaluate_cmd, metric, metric_direction
2. Read .autoresearch/{domain}/{name}/program.md for strategy and constraints
3. Read .autoresearch/{domain}/{name}/results.tsv for experiment history
4. Run: git checkout autoresearch/{domain}/{name}
Then do exactly ONE iteration:
- Review results.tsv: what worked, what failed, what hasn't been tried
- Edit the target file with ONE change (strategy escalation based on run count)
- Commit: git add {target} && git commit -m "experiment: {description}"
- Evaluate: python {skill_path}/scripts/run_experiment.py --experiment {domain}/{name} --single
- Read the output (KEEP/DISCARD/CRASH)
Rules:
- ONE change per experiment
- NEVER modify the evaluator
- If 5 consecutive crashes in results.tsv, delete this cron job (CronDelete) and alert
- After every 10 experiments, update Strategy section of program.md
Current best metric: {read from results.tsv or "no baseline yet"}
Total experiments so far: {count from results.tsv}
Write to .autoresearch/{domain}/{name}/loop.json:
{
"cron_id": "{id from CronCreate}",
"interval": "{user selection}",
"started": "{ISO timestamp}",
"experiment": "{domain}/{name}"
}
Loop started for {domain}/{name}
Interval: {interval description}
Cron ID: {id}
Auto-expires: 3 days (CronCreate limit)
To check progress: /ar:status
To stop the loop: /ar:loop stop {domain}/{name}
Note: Recurring jobs auto-expire after 3 days.
Run /ar:loop again to restart after expiry.
When user runs /ar:loop stop {experiment}:
.autoresearch/{domain}/{name}/loop.json to get the cron IDCronDelete with that IDloop.json/ar:loop to restart. Results persist — the new loop picks up where the old one left off.autoresearch/{domain}/{name}).Weekly Installs
159
Repository
GitHub Stars
6.5K
First Seen
11 days ago
Security Audits
Gen Agent Trust HubPassSocketPassSnykPass
Installed on
codex154
gemini-cli153
kimi-cli152
amp152
github-copilot152
opencode152
Azure Data Explorer (Kusto) 查询技能:KQL数据分析、日志遥测与时间序列处理
125,100 周安装
57 8 1 * * |
monthly |