agentleFS
lingengyuanlingengyuan/my-skills/.claude/skills/wechat-archiver/SKILL.md

wechat-archiver

Skill9 starsChanged 3 days agoMITView on GitHub

Archive WeChat Official Account articles into structured knowledge assets with local Markdown, downloaded images, structured notes, and metadata. Use when archiving, indexing, or saving WeChat articles to an Obsidian vault or knowledge base.

  • Installs packages

The file

Show its rules and text

We split this skill into the rules it gives an agent

14 rules. About 0 are common, 5 are distinctive, and 9 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈5

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 9

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Very specific:

    Determine destination: Verify the target knowledge base directory.

    line 20
  2. Distinctive:

    Fetch article: Run the bundled fetcher tool:

    line 21
  3. Very specific:

    If the fetcher returns STATUS=skipped, the URL and article hash match an existing asset.

    line 25
  4. Very specific:

    Inspect extracted content: Read generated article.md, images/, and meta.json.

    line 26
  5. Very specific:

    Create structured note (note.md): Write a clean analytical note directly in the asset directory:

    line 27
  6. Very specific:

    Update metadata: Ensure meta.json accurately reflects the generated asset paths.

    line 33
  7. Very specific:

    Optional visual artifacts: Create diagram.canvas or table.base only when the user explicitly requests it or when complex spatial/comparison …

    line 34
  8. Very specific:

    Verify output: Verify that article.md, images, note.md, and meta.json exist and pass verification.

    line 35
  9. Distinctive:

    When processing multiple URLs:

    line 51
  10. Distinctive:

    If an article fetch fails, record the error, preserve completed assets, and continue with remaining URLs.

    line 53
  11. Distinctive:

    Do not modify the source inbox list unless directed by the user.

    line 55
  12. Very specific:

    Faithful preservation: Never alter the original article's technical claims or author opinions in article.md.

    line 59
  13. Very specific:

    Minimal optional artifacts: Never create optional diagram.canvas or table.base files without clear informational value.

    line 60
  14. Distinctive:

    Before completing the task, verify:

    line 65

≈ “Common” is an estimate: at least 1 in 5 of the 372,429 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

WeChat Archiver Skill

Archive WeChat Official Account articles into self-contained knowledge asset packages with clean Markdown, local images, a structured analytical note, and verified metadata.

Setup

Python 3.8+ and curl in PATH are required. Install dependencies in the Skill directory:

pip install -r "<skill-dir>/requirements.txt"

Workflow

  1. Determine destination: Verify the target knowledge base directory. Default to outputs/<folder>/<slug>/.
  2. Fetch article: Run the bundled fetcher tool:
    python "<skill-dir>/tools/fetch_article.py" "<article-url>" --output-dir "<target-root>/outputs" --target-folder "<folder>"
    If the fetcher returns STATUS=skipped, the URL and article hash match an existing asset. Verify the existing asset and do not recreate it. Add --refresh only when the user explicitly requests an update.
  3. Inspect extracted content: Read generated article.md, images/, and meta.json.
  4. Create structured note (note.md): Write a clean analytical note directly in the asset directory:
    • 核心摘要:1-2 句直接说明文章解决的问题与核心结论。
    • 关键技术机制与论据:系统原理解析、核心技术决策、支持论据。
    • 可复用资产:可直接提取的代码、配置、命令或规则。
    • 局限与边界:原文适用的范围、未解决的问题或作者主观偏见。
    • 绝不编造泛泛而谈的寒暄、大而无当的空话或虚假深刻的升华。
  5. Update metadata: Ensure meta.json accurately reflects the generated asset paths.
  6. Optional visual artifacts: Create diagram.canvas or table.base only when the user explicitly requests it or when complex spatial/comparison relationships provide clear visual value.
  7. Verify output: Verify that article.md, images, note.md, and meta.json exist and pass verification.

Output Package Structure

<target-root>/outputs/<folder>/<slug>/
├── article.md          # Faithful Markdown of original article
├── images/             # Downloaded local images (if present)
├── note.md             # Structured analytical note
├── meta.json           # Machine-readable asset metadata (title, author, url, asset_id)
├── diagram.canvas      # Optional (only when explicitly requested)
└── table.base          # Optional (only when explicitly requested)

Batch Processing

When processing multiple URLs:

  1. Process articles sequentially using the standard single-article workflow.
  2. If an article fetch fails, record the error, preserve completed assets, and continue with remaining URLs.
  3. Report a final summary listing successful paths and failed URLs.
  4. Do not modify the source inbox list unless directed by the user.

Guidelines & Boundaries

  • Faithful preservation: Never alter the original article's technical claims or author opinions in article.md.
  • Minimal optional artifacts: Never create optional diagram.canvas or table.base files without clear informational value.
  • Completion definition: An archive is complete only when article.md, downloaded images, note.md, and meta.json exist and pass verification.

Verification Checklist

Before completing the task, verify:

  1. article.md exists and image references link to local files in images/.
  2. Local images exist on disk with non-zero file sizes.
  3. note.md contains verified analytical sections and source attribution without AI fluff.
  4. meta.json contains accurate artifact paths and metadata.
  5. Output reports the exact asset directory path to the user.

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.