← Back to Skills Marketplace
besty0121

Experiment Notes

by besty0121 · GitHub ↗ · v1.0.0 · MIT-0
cross-platform ✓ Security Clean
117
Downloads
0
Stars
0
Active Installs
1
Versions
Install in OpenClaw
/install experiment-notes
Description
Track, search, and learn from experiments. Automatic logging of trial-and-error, success/failure patterns, and distilled lessons. Prevents repeating mistakes.
README (SKILL.md)

实验笔记系统 / Experiment Notes

让 Agent 从经验中学习,不再重复犯错。

Every agent fails. Smart agents remember why.

核心理念

普通 Agent:

试了 → 失败了 → 忘了 → 下次又试 → 又失败了

用了实验笔记的 Agent:

试了 → 失败了 → 记下来了 → 下次查一下 → 直接用对的方案

安装

skill 安装后,数据存储在 ~/.openclaw/memory/experiments/

  • experiments.jsonl — 所有实验记录
  • lessons.jsonl — 提炼的最佳实践

首次使用会自动创建目录。

命令参考

所有命令通过 python \x3Cskill_dir>/scripts/expnote.py 调用。

记录实验

python expnote.py log \
  --task "从 Docker Hub 拉取镜像" \
  --outcome failure \
  --tags docker,network \
  --cmd "docker pull nginx" \
  --error "Error response from daemon: Get https://registry-1.docker.io/v2/: net/http: request canceled" \
  --fix "配置 Docker 镜像加速器" \
  --lesson "国内环境拉 Docker 镜像大概率超时,先配加速器"

outcome 取值:

  • success — 成功了 ✅
  • failure — 失败了 ❌
  • partial — 部分成功 ⚠️

搜索历史

# 全文搜索
python expnote.py search "docker"

# 限制结果数
python expnote.py search "clawhub publish" --limit 5

查找类似任务

在开始新任务前,先查查之前有没有做过类似的:

python expnote.py similar "发布 skill 到 ClawHub"

查看经验教训

# 所有教训
python expnote.py lessons

# 按标签过滤
python expnote.py lessons --tag docker

统计概览

python expnote.py stats

显示:总实验数、成功率/失败率、热门标签、最近活动。

提炼最佳实践

从多次实验中总结通用经验:

python expnote.py distill \
  --tags clawhub,publish \
  --lesson "clawhub publish 必须显式指定 --version,不支持自动递增"

给 Agent 的使用指南

在你的 AGENTS.md 中添加以下规则:

## 实验笔记

每次尝试新操作时:

1. **之前**:运行 `expnote.py similar "任务描述"`,看看有没有历史经验
2. **之后**:无论成功失败,运行 `expnote.py log` 记录结果
3. **定期**:review 实验记录,用 `expnote.py distill` 提炼通用教训

记录格式:
- task: 简短描述(动词开头)
- outcome: success / failure / partial
- tags: 相关技术/工具(逗号分隔)
- cmd: 实际执行的命令
- error: 报错信息(失败时)
- fix: 怎么修复的
- lesson: 一句话总结(最重要!)

数据格式

每条实验记录(experiments.jsonl 中的一行):

{
  "id": "exp_a1b2c3d4e5f6",
  "timestamp": "2026-04-02T08:58:00+08:00",
  "task": "发布 ClawHub skill",
  "outcome": "partial",
  "attempt": {
    "command": "clawhub publish ./skill --no-input",
    "description": "发布小红书浏览器 skill"
  },
  "result": {
    "error": "Error: --version must be valid semver",
    "fix": "加上 --version 1.1.0"
  },
  "tags": ["clawhub", "publish", "skill"],
  "lesson": "clawhub publish 必须显式指定 --version"
}

目录结构

experiment-notes/
├── SKILL.md              # 本文件
├── scripts/
│   └── expnote.py        # CLI 工具
└── templates/
    └── agents-snippet.md  # AGENTS.md 集成片段

设计理念

  • JSONL 格式:追加写入,不怕并发,方便 grep
  • 纯文件存储:不需要数据库,不需要服务端
  • 标签系统:多维分类,支持按领域过滤
  • 提炼机制:不只是记录,还要总结成可复用的知识

局限性

  • 搜索是关键词匹配,不是语义搜索(够用但不完美)
  • 跨 agent 共享需要手动同步文件
  • 没有自动清理机制(可以手动删除旧记录)
Usage Guidance
This skill appears to do what it says and stores notes locally under ~/.openclaw/memory/experiments. Before installing: (1) understand that 'automatic logging' is not implemented as a background daemon — the agent or user must invoke the CLI or add the provided AGENTS.md rules to call it; (2) avoid logging secrets or credentials (commands/errors may contain tokens or passwords); (3) consider setting restrictive file permissions or encrypting the directory if experiment logs may contain sensitive data; (4) if you will allow autonomous agent invocation, ensure the agent is trusted so it doesn't log or leak sensitive commands automatically.
Capability Analysis
Type: OpenClaw Skill Name: experiment-notes Version: 1.0.0 The 'experiment-notes' skill is a productivity tool designed to help AI agents log, search, and learn from their task history. The core logic in `scripts/expnote.py` manages local JSONL files within `~/.openclaw/memory/experiments/` using standard Python libraries. There is no evidence of network activity, unauthorized file access, shell execution, or malicious prompt injection instructions in `SKILL.md` or the associated templates.
Capability Assessment
Purpose & Capability
Name/description (experiment notes, lessons, search) align with the provided CLI and files. The included script implements logging, search, similarity, lessons, stats and distill features, which match the stated purpose.
Instruction Scope
SKILL.md describes 'automatic logging' in the description, but the runtime instructions and CLI are manual (agent must run expnote.py or be configured to call it). Instructions do not request reading unrelated files or credentials. Note: entries (cmd, error, fix) are free-text — users/agents could log sensitive strings if care is not taken.
Install Mechanism
No install spec; this is an instruction-only skill with a bundled Python script. No downloads or external packages are required beyond a local python binary.
Credentials
The skill declares no required env vars or credentials. The code only uses Path.home() to place files under ~/.openclaw/memory/experiments. No network calls, third-party APIs, or unrelated credentials are requested.
Persistence & Privilege
always:false and no install hooks. The skill persists data locally in its own directory but does not modify other skills or global agent settings. Model invocation is allowed by default (normal) but not combined with broad privileges.
How to Use
  1. Make sure OpenClaw is installed (local or Docker)
  2. Run the install command in chat: /install experiment-notes
  3. After installation, invoke the skill by name or use /experiment-notes
  4. Provide required inputs per the skill's parameter spec and get structured output
Version History
v1.0.0
experiment-notes v1.0.0 - Initial release of the experiment-notes skill. - Provides CLI commands to log, search, and distill lessons from experiments. - Automatically tracks trial outcomes, fixes, and distilled best practices to prevent repeated mistakes. - Stores all data locally in JSONL files for easy access and management. - Supports tag-based filtering, statistics overview, and guidance for agent integration.
Metadata
Slug experiment-notes
Version 1.0.0
License MIT-0
All-time Installs 0
Active Installs 0
Total Versions 1
Frequently Asked Questions

What is Experiment Notes?

Track, search, and learn from experiments. Automatic logging of trial-and-error, success/failure patterns, and distilled lessons. Prevents repeating mistakes. It is an AI Agent Skill for Claude Code / OpenClaw, with 117 downloads so far.

How do I install Experiment Notes?

Run "/install experiment-notes" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.

Is Experiment Notes free?

Yes, Experiment Notes is completely free, licensed under MIT-0. You can download, install and use it at no cost.

Which platforms does Experiment Notes support?

Experiment Notes is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).

Who created Experiment Notes?

It is built and maintained by besty0121 (@besty0121); the current version is v1.0.0.

💬 Comments