The web page archiver is working! Key features:

✅ HTML-to-SHM conversion with hierarchy inference

✅ Image upload to IPFS

✅ Dedicated archive keys (one per source)

✅ Archive profile creation with metadata

✅ Provenance tracking (source URL, timestamp, tool)

Dedicated archive account with its own cryptographic identity:

• Mozilla Monitor Plus article - First archived article with images

GitHub Markdown Guide - 160 blocks, proper hierarchy

# Create a dedicated archive identity
shm-archive create-archive web "example-site" \
  --description "Archived articles from Example" \
  --source-url "https://example.com"

# Archive a URL to that identity
shm-archive url "https://example.com/article" \
  --key archive-web-example-site

# Or test parsing without publishing
shm-archive test-parse "https://example.com/article"

The hardest problem: HTML is flat, SHM requires tree structure. Solved by tracking heading levels with a stack algorithm.

⬜ Wikipedia archiver (MediaWiki API)

⬜ Social media (Twitter, Mastodon)

⬜ Web UI for triggering archives

⬜ Batch archiving support

Tool: ~/shm-web-archiver/

Key files:

• src/html-to-shm.js - HTML parsing and hierarchy inference

• src/image-uploader.js - IPFS image upload

• src/archive-key.js - Dedicated key management

• bin/archive.js - CLI interface