Bookmark Ninja

Bookmark Ninja

This skill parses Netscape HTML bookmark exports (the universal format produced by Chrome, Firefox, Edge, and Brave) and produces structured JSON or CSV indexes ready for agent consumption. All processing is handled by scripts/bookmark-parser.py. The agent's role is to recognize when a bookmark task applies and invoke the script with the appropriate flags — no API keys, no external services, no configuration.

When to Use This Skill

• Convert browser bookmarks to agent-queryable format — transform an exported HTML file into a structured JSON or CSV index
• Build an incremental knowledge base — merge periodic bookmark exports into a growing index without losing existing data
• Organize accumulated research — preserve full folder hierarchy as category breadcrumb paths (e.g., Research > AI > Papers)
• Verify URL liveness — identify dead links via HEAD request checks before querying
• Merge multiple bookmark sources — combine exports from different browsers or time periods with conflict resolution
• Archive bookmarks before browser migration — snapshot current bookmarks as a portable index before reorganizing

How to Run the Script

The script lives at scripts/bookmark-parser.py relative to the skill folder. Use the path appropriate to the runtime environment.

Basic parse — reads an HTML export and writes bookmarks-index.json:

python3 scripts/bookmark-parser.py bookmarks.html

Custom output path — write the index to a specific location:

python3 scripts/bookmark-parser.py bookmarks.html -o ~/research/sources.json

Merge with existing index — add new bookmarks without overwriting; prompts interactively on conflicts:

python3 scripts/bookmark-parser.py new-bookmarks.html -o index.json --merge

Auto-resolve merge conflicts — skip interactive prompts by keeping old or new data:

# Keep the old entry on conflict
python3 scripts/bookmark-parser.py new-bookmarks.html -o index.json --merge --keep-old

# Keep the new entry on conflict
python3 scripts/bookmark-parser.py new-bookmarks.html -o index.json --merge --keep-new

URL liveness check — adds "alive": true/false to each entry (requires requests; note: slow on large collections):

python3 scripts/bookmark-parser.py bookmarks.html --check-alive -o verified.json

Statistics only — print a summary without writing any output file:

python3 scripts/bookmark-parser.py bookmarks.html --stats

CSV output — produce a CSV file instead of JSON:

python3 scripts/bookmark-parser.py bookmarks.html --format csv

Both formats at once:

python3 scripts/bookmark-parser.py bookmarks.html --format both

Output Format

Each entry in the JSON array contains seven fields:

Example entry:

{
  "url": "https://dehashed.com/",
  "title": "DeHashed — #FreeThePassword",
  "category": "OSINT > Breach Data/Dumps",
  "description": "Searches breach, paste, or leak data for emails, usernames, and exposed records.",
  "date_added": "2026-03-14T09:20:00",
  "icon": "",
  "alive": true
}

For a multi-entry example, see references/example-output.json.

Agent Query Patterns

import json

with open("bookmarks-index.json") as f:
    bookmarks = json.load(f)

# Search by keyword (title or description)
keyword = "machine learning"
matches = [b for b in bookmarks if keyword.lower() in b["title"].lower()
           or keyword.lower() in b.get("description", "").lower()]

# Filter by category
email_tools = [b for b in bookmarks if "Email Search" in b["category"]]

# List all categories
categories = sorted(set(b["category"] for b in bookmarks))

# Find dead links (requires --check-alive run)
dead = [b for b in bookmarks if b.get("alive") == False]

# Recent bookmarks (last 30 days)
from datetime import datetime, timedelta
cutoff = datetime.now() - timedelta(days=30)
recent = [b for b in bookmarks
          if b["date_added"] and datetime.fromisoformat(b["date_added"]) > cutoff]

Command-Line Reference

usage: bookmark-parser.py [-h] [-o OUTPUT] [--format {json,csv,both}]
                          [--merge] [--keep-old] [--keep-new]
                          [--check-alive] [--stats]
                          input

Convert browser bookmarks to machine-readable index

positional arguments:
  input                 HTML bookmark file to parse

options:
  -h, --help            show this help message and exit
  -o OUTPUT, --output OUTPUT
                        Output file path (default: bookmarks-index.json)
  --format {json,csv,both}
                        Output format (default: json)
  --merge               Merge with existing index file
  --keep-old            On merge conflict, keep old entry (default: prompt)
  --keep-new            On merge conflict, keep new entry (default: prompt)
  --check-alive         Check URL liveness via HEAD request (adds latency)
  --stats               Print statistics only, don't save

Workflow Example

1. Initial import — parse your first export:

   python3 scripts/bookmark-parser.py research-bookmarks.html -o research-library.json
   

2. Monthly merges — add each new export without losing history:

   python3 scripts/bookmark-parser.py bookmarks-june.html -o research-library.json --merge --keep-new
   

3. Quarterly liveness check — flag dead links before querying:

   python3 scripts/bookmark-parser.py bookmarks-current.html --check-alive -o research-library.json --merge --keep-new
   

4. Agent queries the index — load the JSON and filter programmatically using the patterns in the Agent Query Patterns section above.

Troubleshooting

"requests library not available" warning Liveness check requires requests. Install with pip install requests.

Encoding errors on import The parser uses errors="ignore" for malformed HTML. If issues persist, re-export from the browser.

Category hierarchy looks wrong Bookmark exports nest categories via <DL> tags. Ensure the export includes folder structure (not a flat list). In Chrome/Firefox, verify that folder structure is preserved during export.

Merge conflicts on every entry Usually caused by date/time format changes between exports. Use --keep-new to auto-accept updated metadata without prompting.