experiment-watchdog SKILL.md skillA
experiment-watchdog from yichen-zju/claude-for-research is an agent skill: a SKILL.md folder an AI agent loads when a task calls for it.
mdr add yichen-zju/claude-for-research/experiment-watchdog@git:20260930.9b33ccfPins git:20260930.9b33ccf in mdr.lock. Pin the fingerprint instead.
- Audit grade
- A, 17 of 17 checks passed
- Kind
- SKILL.md skill
- Repository
- yichen-zju/claude-for-research
- Path
skills/experiment-watchdog/SKILL.md- Versions recorded
- 1
- Latest version
- git:20260930.9b33ccf, committed Sep 30, 2026
- Size
- 1,960 bytes
- Repository stars
- 10
- License
- none declared
Server-side watchdog daemon that monitors registered training/download/loop tasks and surfaces DEAD/STALLED/IDLE anomalies. Use when running unattended or overnight experiments and you want silent deaths or stalls to be detected instead of discovered hours later. Wraps the ARIS watchdog.py (stdlib-only).
Collected from public GitHub. Every version is saved exactly as it was and never changes. Install it at an exact version with the mdr tool, and every file is checked before it reaches your agent. The main file is checked against the SHA-256 fingerprint recorded here, and the others against the fingerprints GitHub keeps for its source commit. The automatic checks below grade the latest version, and the same checks always give the same file the same grade.
What the file says
# Experiment Watchdog
来源：ARIS（MIT License）的 `watchdog.py`，stdlib 单文件，无需安装依赖。
**原则：watchdog 只检测，不裁判。** 它能说"这个任务卡死了/进程没了"，永远不说"这个结果够好"。
## 用法
```bash
W=<skill-dir>/scripts/watchdog.py
# 1. 启动守护进程（前台运行，用 tmux/screen 挂后台）
python3 $W --interval 60 & # 每 60 秒巡检一次
# 2. 注册训练任务（screen 会话里的训练）
python3 $W --register '{"name":"exp01","type":"training","session":"exp01","session_type":"screen","gpus":[0,1]}'
# 3. 注册下载任务
python3 $W --register '{"name":"dl01","type":"download","session":"dl01","session_type":"tmux","target_path":"/path/to/file"}'
# 4. 随时看状态（汇总，一行一个任务）
python3 $W --status
# 5. 任务结束注销
python3 $W --unregister exp01
```
## 输出
- `/tmp/aris-watchdog/status/summary.txt` —— 每任务一行，适合 `/loop` 或 CronCreate 定时读取
- `/tmp/aris-watchdog/alerts.log` —— DEAD/STALLED/IDLE 异常日志，跨会话恢复时先读这个
- `/tmp/aris-watchdog/status/<task>.json` —— 单任务详细状态
## 与工作流配合
- 跑 `experiment-queue` 或过夜实验时：启动 watchdog + 注册任务，然后可以用 CronCreate 每 10-30 分钟读一次 `summary.txt`，有异常再深入。
- 无人值守循环（karpathy 模式）开始前注册为 `loop` 类型任务 —— 静默死亡会在下次巡检时暴露为 STALE，而不是几小时后才被你发现。Read the whole file at its exact version.
How to install
mdr add yichen-zju/claude-for-research/experiment-watchdog@sha256:802a60667c18896cThe line at the top of this page pins the label git:20260930.9b33ccf, which is what most people want. This one pins the sha256 fingerprint of the contents instead, so it survives a label being moved. Either way the exact version is written to mdr.lock, and mdr install downloads and checks those contents again, so it installs them exactly or stops with an error.
[](https://markdownregistry.com/a/art_ehgs4oeabl4ibykw)
2 badge image fetches in 30 days. What the badge shows.
Versions
The registry has recorded 1 version of this file, committed on Sep 30, 2026.
Audit of the latest version
- pass: Frontmatter block present
- pass: Frontmatter declares a name
- pass: Frontmatter declares a description
- pass: Size between 200 bytes and 200 KB (1960 bytes)
- pass: No zero-width or bidi control characters
- pass: No instruction hidden inside an HTML comment
- pass: No link to an exfiltration or paste host
- pass: No credential-shaped string
- pass: No instruction to send local credentials anywhere
- pass: No text hidden with inline styles
- pass: No prompt-injection phrasing
- pass: No curl or wget piped into a shell
- pass: No recursive delete of root, home or parent
- pass: No instruction to read or print local credentials
- pass: No base64 blob over 200 characters
- pass: No link to a raw IP address
- pass: No script tag
Source
yichen-zju/claude-for-research · 10 stars · license none declared · pushed 2026-10-21 · branch main
API
GET https://markdownregistry.com/api/v1/artifacts/art_ehgs4oeabl4ibykw GET https://markdownregistry.com/api/v1/resolve?ref=yichen-zju/claude-for-research/experiment-watchdog GET https://markdownregistry.com/api/v1/blob/802a60667c18896cff93c4c74f7c2abecc5c98cd5e9d2c63d40a3421d4ac3b03
Your AI assistant does the legwork. You hear about the deals worth your word. Hand yours the standing instructions at modelranch.com and it joins the network that reads files like this one.
Guides: the SKILL.md format and its required fields · what to check before you install a skill · how to pin a skill to an exact version · where to find agent skills · keeping a team on the same skills · which agent file does what.
More from yichen-zju/claude-for-research
Every file in yichen-zju/claude-for-research