dojo.md
University for AI agents. 92 courses, 4400+ scenarios, any model via OpenRouter. Auto-training loops generate per-model SKILL.md documents. Works with Claude Code, OpenClaw, Cursor, Windsurf. No fine-tuning req...
Privacy-first re-crawl and diff workflow for Tencent Docs (docs.qq.com) update checks. Use when asked to re-crawl Tencent Docs links, verify whether content changed since last snapshot, and produce a structured update report. Supports sheet/doc/aio links and snapshot comparison.
请根据来源仓库和你的 AI 工具说明安装此技能。先阅读技能文件和依赖要求,再添加到工作流。
SkillHub CLI
skillhub install chinasilva/tencent-doc-update-watch打开来源仓库,查看说明和兼容环境后,按来源提供的命令或步骤安装。
将来源链接提供给你的 Agent,让它先读取说明、核对依赖,并向你说明将执行的安装步骤。
Use this skill to re-crawl Tencent Docs links and detect updates with snapshot diffing. The workflow outputs:
manifest.json (structured machine-readable result)report.md (human-readable diff summary)Run with default doc list:
python scripts/check-qq-doc-updates.py \
--config references/default-docs.json \
--workspace /tmp/tencent-doc-watch
Run with explicit baseline:
python scripts/check-qq-doc-updates.py \
--config /abs/path/my-docs.json \
--workspace /tmp/tencent-doc-watch \
--compare /tmp/tencent-doc-watch/snapshots/20260305_101500/manifest.json
Default behavior is privacy-first:
opendoc_url fields.Keep raw files only when explicitly needed for debugging:
python scripts/check-qq-doc-updates.py \
--config references/default-docs.json \
--workspace /tmp/tencent-doc-watch \
--keep-raw
Use JSON:
{
"docs": [
{
"name": "Example Sheet",
"url": "https://docs.qq.com/sheet/DEXAMPLE_SHEET_001?tab=sample"
},
{
"name": "Example Doc",
"url": "https://docs.qq.com/doc/DEXAMPLE_DOC_001"
}
]
}
Optional field:
id: Tencent doc ID (auto-parsed from URL when omitted)Generated under <workspace>/snapshots/<label>/:
manifest.json
report.md
raw/ only when --keep-raw is enabledReport status values:
UNCHANGED
CHANGED
NEW
FIRST_RUN
Comparison currently checks:
last_modify_ms
rev
cgi_code
title
pad_type
此 Skill 固化“腾讯文档重新抓取 + 与历史快照对比”的流程。默认开启隐私最小化策略:
--keep-raw)opendoc_url 查询参数manifest.json 中移除页面...这些信息来自收录时的来源分析,可能滞后;使用前请核对最新来源文件。
# Tencent Doc Update Watch
## Overview
Use this skill to re-crawl Tencent Docs links and detect updates with snapshot diffing.
The workflow outputs:
- `manifest.json` (structured machine-readable result)
- `report.md` (human-readable diff summary)
## Quick Run
Run with default doc list:
```bash
python scripts/check-qq-doc-updates.py \
--config references/default-docs.json \
--workspace /tmp/tencent-doc-watch
```
Run with explicit baseline:
```bash
python scripts/check-qq-doc-updates.py \
--config /abs/path/my-docs.json \
--workspace /tmp/tencent-doc-watch \
--compare /tmp/tencent-doc-watch/snapshots/20260305_101500/manifest.json
```
## Privacy Defaults
Default behavior is privacy-first:
- Raw HTML/opendoc/cookie files are removed after parsing.
- URL query values are redacted in `opendoc_url` fields.
- Public URL in manifest removes query strings to avoid leaking tokens.
Keep raw files only when explicitly needed for debugging:
```bash
python scripts/check-qq-doc-updates.py \
--config references/default-docs.json \
--workspace /tmp/tencent-doc-watch \
--keep-raw
```
## Config Format
Use JSON:
```json
{
"docs": [
{
"name": "Example Sheet",
"url": "https://docs.qq.com/sheet/DEXAMPLE_SHEET_001?tab=sample"
},
{
"name": "Example Doc",
"url": "https://docs.qq.com/doc/DEXAMPLE_DOC_001"
}
]
}
```
Optional field:
- `id`: Tencent doc ID (auto-parsed from URL when omitted)
## Output and Status
Generated under `<workspace>/snapshots/<label>/`:
- `manifest.json`
- `report.md`
- `raw/` only when `--keep-raw` is enabled
Report status values:
- `UNCHANGED`
- `CHANGED`
- `NEW`
- `FIRST_RUN`
## Key Detection Signals
Comparison currently checks:
- `last_modify_ms`
- `rev`
- `cgi_code`
- `title`
- `pad_type`
## 中文补充
此 Skill 固化“腾讯文档重新抓取 + 与历史快照对比”的流程。默认开启隐私最小化策略:
- 不持久化原始抓取内容(除非加 `--keep-raw`)
- 脱敏 `opendoc_url` 查询参数
- 在 `manifest.json` 中移除页面...University for AI agents. 92 courses, 4400+ scenarios, any model via OpenRouter. Auto-training loops generate per-model SKILL.md documents. Works with Claude Code, OpenClaw, Cursor, Windsurf. No fine-tuning req...
OpenClaw Skill for stock data analysis
Generative Engine Optimization (GEO) for AI search visibility. Optimize content to appear in ChatGPT, Perplexity, Claude, and Google AI Overviews. Use when optimizing websites, pages, or content for LLM discove...
macOS WPS Office workflow helper skill for safer document preparation, conversion, export, and compatibility guidance