archivist

name	archivist
description	Universal content cataloging system. Save URLs, files, snippets, or voice notes with descriptions, auto-tag everything, make it all searchable. Use when: building a personal knowledge base, saving reference material, or organizing research. NOT for: file backup, website mirroring, or large-scale web scraping.
homepage	https://pawhub.ai/archivist
metadata	{"openpaw":{"emoji":"📁","requires":{"bins":["curl","sqlite3"]},"install":[{"id":"npm","kind":"node","package":"@pawhub/archivist","bins":["archivist"],"label":"Install Archivist (npm)"}]}}

Archivist 📁

Your personal librarian. Save anything — URLs, files, notes, screenshots — and Paw makes it searchable, tagged, and instantly retrievable.

When to Use

✅ USE this skill when:

• "Save this article for later"
• Building a personal knowledge base
• Organizing research materials
• Archiving reference links
• Saving code snippets with context
• Creating a searchable bookmark library
• Capturing voice notes with transcription
• Organizing screenshots with descriptions
• Building a "second brain"

When NOT to Use

❌ DON'T use this skill when:

• Full file backup/sync → use Time Machine, Backblaze, etc.
• Website mirroring → use wget, HTTrack
• Large-scale web scraping → use Scrapy, Apify
• Photo library management → use Photos, Lightroom
• Document version control → use Git, Dropbox versioning
• Collaborative wikis → use Notion, Obsidian Sync

Setup

Install Archivist:

npm install -g @pawhub/archivist

Initialize your archive:

# Create archive database
archivist init

# Set storage location (default: ~/.archivist)
archivist config set storage-path ~/Documents/Archive

# Verify setup
archivist status

Example Interactions

Saving Content

You: Save this for later: https://example.com/great-article

Paw: fetches and archives the page

✅ Saved: "How to Build Better APIs"

Captured:

• Full HTML (for offline reading)
• Screenshot (visual reference)
• Extracted text (for search)
• Title: "How to Build Better APIs"
• Author: Jane Smith
• Published: 2026-01-15

Auto-tags: #api-design #programming #webdev

Summary (AI-generated):

Article discusses REST vs GraphQL trade-offs, API versioning strategies, and authentication best practices. Includes code examples in Python and Node.js.

Archive ID: arch_abc123

Want to add custom tags or notes? Just say the word. 🐾

---

You: Add tags: must-read, reference, backend

Paw: updates the entry

✅ Tags updated:

• #api-design (auto)
• #programming (auto)
• #webdev (auto)
• #must-read (you)
• #reference (you)
• #backend (you)

Searchable immediately. Try: archivist search "api design must-read" 📚

---

You: Save this snippet:

def retry_with_backoff(func, max_retries=3):
    for attempt in range(max_retries):
        try:
            return func()
        except Exception as e:
            if attempt == max_retries - 1:
                raise
            time.sleep(2 ** attempt)

Paw: analyzes and archives the code

✅ Saved: Code snippet

Details:

• Language: Python
• Type: Function definition
• Name: retry_with_backoff
• Lines: 9

Auto-tags: #python #error-handling #retry-logic

AI Summary:

Implements exponential backoff retry logic. Useful for API calls and network operations.

Archive ID: arch_xyz789

Suggested title: "Python Retry with Exponential Backoff"

Looks good? (yes/edit)

---

You: Yes

Paw: Perfect. Archived under #python #error-handling. Search it anytime with archivist search retry or just ask me "find that Python retry function." 🔍

---

Searching Your Archive

You: Find articles about API design

Paw: searches archive

Found 4 results for "API design":

1. How to Build Better APIs (2 days ago) 📄 Article · https://example.com/great-article Tags: #api-design #programming #must-read REST vs GraphQL trade-offs, versioning...

2. REST API Best Practices 2026 (1 week ago) 📄 Article · https://dev.to/rest-best-practices Tags: #api-design #rest #backend Modern patterns for REST APIs including...

3. Designing APIs for Scale (2 weeks ago) 📄 Article · https://scaling.example.com Tags: #api-design #distributed-systems Load balancing, rate limiting, caching strategies...

4. GraphQL Schema Design (3 weeks ago) 📄 Article · https://graphql-guide.example.com Tags: #api-design #graphql Schema-first development, resolvers...

Want me to open one? (1-4) 📖

---

You: Open 1

Paw: opens the archived page

Opening "How to Build Better APIs" in your browser... (Archived version — works offline)

Quick actions:

• archivist note arch_abc123 "Used this for the v2 API redesign"
• archivist share arch_abc123 (get shareable link)
• archivist export arch_abc123 --format pdf

Anything else? 🐾

---

Voice Notes

You: [voice message] "Idea for the new feature: what if we added real-time collaboration like Figma but for code? Users could see each other's cursors and edits live. Would need WebSockets and operational transforms. Check out Yjs library."

Paw: transcribes and archives

✅ Voice note saved

Transcription:

Idea for the new feature: what if we added real-time collaboration like Figma but for code? Users could see each other's cursors and edits live. Would need WebSockets and operational transforms. Check out Yjs library.

Auto-tags: #ideas #real-time #collaboration #websockets

Extracted mentions:

• Figma (reference)
• Yjs (library to research)
• WebSockets (tech)
• Operational transforms (concept)

AI Summary:

Feature idea: real-time collaborative code editing with live cursors. Implementation notes: WebSockets + operational transforms. Research Yjs library.

Archive ID: arch_voice_456

Created follow-up todo: "Research Yjs library for operational transforms"

Want me to find similar ideas in your archive? 💡

---

Advanced Search

You: Show me everything I saved last month about Python

Paw: filters archive

Last month (Jan 2026) · Python: 12 items

Breakdown:

• 🔗 Articles: 7
• 📄 Code snippets: 4
• 🎤 Voice notes: 1

Top tags:

• #python (12)
• #async (5)
• #fastapi (3)
• #testing (3)

Most referenced:

1. "Async Python Patterns" (opened 8 times)
2. "FastAPI Best Practices" (opened 5 times)

Want the full list or just the highlights? 📊

---

You: Just highlights

Paw: Top 3 from last month:

1. Async Python Patterns 📄 Article · Saved Jan 5 Tags: #python #async #advanced Comprehensive guide to asyncio, context managers, generators Why it stood out: You've opened this 8 times

2. Python retry function snippet 📋 Code · Saved Jan 12 Tags: #python #error-handling Exponential backoff retry logic Why it stood out: You added note "use this everywhere"

3. Testing async code 📄 Article · Saved Jan 18 Tags: #python #testing #async pytest-asyncio examples and best practices Why it stood out: Tagged as #must-read

Want to review any of these? 🐾

---

Organization

You: How's my archive looking?

Paw: generates statistics

📁 Your Archive

Total items: 847 Added this week: 23 Storage used: 1.2 GB

Breakdown by type:

• 🔗 URLs/Articles: 512 (60%)
• 📄 Code snippets: 178 (21%)
• 📎 Files: 94 (11%)
• 🎤 Voice notes: 41 (5%)
• 📸 Screenshots: 22 (3%)

Most common tags:

1. #programming (234)
2. #javascript (142)
3. #reference (128)
4. #api-design (89)
5. #must-read (67)

Recent additions:

• Today: 3 items
• Yesterday: 7 items
• This week: 23 items

Search hotspots:

• You search for "api" most often (42 times)
• "async" is trending up (15 searches this month)

Suggestions:

• 23 items have no tags → want me to auto-tag?
• 12 broken links detected → run health check?
• 8 duplicate URLs found → merge them?

Need anything? 📚

Commands

Saving Content

# Save a URL
archivist save "https://example.com/article"

# Save with custom title and tags
archivist save "https://example.com" \
  --title "Great Article" \
  --tags "programming,reference"

# Save a file
archivist save-file ~/Downloads/paper.pdf \
  --tags "research,ml"

# Save a code snippet
archivist save-snippet code.py \
  --language python \
  --tags "async,example"

# Save from clipboard
archivist save-clipboard --tags "quick-save"

# Voice note (with transcription)
archivist voice-note record --tags "ideas"

Searching

# Full-text search
archivist search "api design patterns"

# Search by tag
archivist search --tag python --tag async

# Search by type
archivist search --type article "REST"

# Search by date
archivist search --since "2026-01" --until "2026-02"

# Fuzzy search (typo-tolerant)
archivist search "pyton asyncio" --fuzzy

# Show recent additions
archivist recent --limit 10

Organization

# List all tags
archivist tags list

# Rename a tag
archivist tags rename old-tag new-tag

# Merge tags
archivist tags merge python,python3 --into python

# Auto-tag untagged items
archivist auto-tag

# Find duplicates
archivist duplicates --action review

# Health check (broken links, etc.)
archivist health-check --fix

Exporting

# Export single item
archivist export arch_abc123 --format pdf

# Export by tag
archivist export --tag must-read --format markdown

# Export search results
archivist search "api" --export --format html

# Backup entire archive
archivist backup ~/Backups/archivist-backup.zip

# Generate static site
archivist generate-site --output ~/Sites/archive

Sharing

# Generate shareable link
archivist share arch_abc123

# Share with expiration
archivist share arch_abc123 --expires 7d

# Share collection by tag
archivist share-collection --tag python-resources

# Generate QR code for mobile sharing
archivist share arch_abc123 --qr

Telegram Integration

Once connected, you can save from anywhere:

You: [forwards article link]

Paw: Want me to archive this?

URL: https://example.com/great-article Title: "How to Build Better APIs"

React with 📚 to save, or reply with tags.

---

You: 📚 api-design must-read

Paw: ✅ Archived with tags: #api-design #must-read

Searchable via /archive search api design 🐾

---

Search from Telegram:

You: /archive search python async

Paw: Found 7 results. Top 3:

1. Async Python Patterns (Jan 5)
2. Python asyncio guide (Dec 18)
3. FastAPI async examples (Jan 12)

Reply with a number to view, or /archive search python async --all for full list.

Smart Features

Auto-tagging

Archivist analyzes content and suggests tags:

# Enable auto-tagging
archivist config set auto-tag true

# Set confidence threshold (0.0-1.0)
archivist config set auto-tag-confidence 0.7

# Review auto-tag suggestions
archivist review-suggestions

Summaries

AI-generated summaries for every saved item:

# Generate summary for existing item
archivist summarize arch_abc123

# Auto-summarize on save
archivist config set auto-summarize true

# Adjust summary length
archivist config set summary-length "brief" # brief, medium, detailed

Related Content

Find connections between archived items:

# Show related items
archivist related arch_abc123

# Build knowledge graph
archivist graph --tag programming

# Find connections
archivist connect "api design" "authentication"

Configuration

# Show config
archivist config show

# Storage location
archivist config set storage-path ~/Documents/Archive

# Database location
archivist config set db-path ~/.archivist/archive.db

# Screenshot quality
archivist config set screenshot-quality high # low, medium, high

# Full-text extraction
archivist config set extract-text true

# Offline mode (save full HTML)
archivist config set offline-mode true

# Auto-summarize
archivist config set auto-summarize true

# Language for voice transcription
archivist config set transcription-language en

Storage & Performance

Disk usage:

• Articles: ~500KB each (HTML + screenshot)
• Code snippets: ~10KB each
• Files: original size preserved
• Voice notes: audio + text (~1MB each)

Database:

• SQLite (lightweight, portable)
• Full-text search via FTS5
• Typical size: ~50MB for 1000 items

Performance:

• Search: <100ms for 10,000 items
• Save URL: ~2-5 seconds (fetch + process)
• Voice transcription: ~3-10 seconds

Tips from Paw

"Tag as you go. It's way easier than bulk-tagging 500 items later. Trust me."

"Voice notes are underrated. Capture ideas immediately, let me transcribe and tag them overnight."

"The AI summaries are legit. Enable auto-summarize and you'll actually remember why you saved things."

"Use the Telegram integration. Forward links directly from any app. Zero friction = more saving = better knowledge base."

"Run archivist health-check monthly. Broken links pile up fast."

Pricing

• Free tier: 100 items, basic search, manual tagging
• Pro: $8/month — unlimited items, AI tagging/summaries, voice transcription, advanced search
• Team: $25/month — shared archives, collaboration features, API access

Install from PawHub or pawhub.ai/archivist

Privacy

• All data stored locally on your OpenPaw gateway
• Optional cloud sync (encrypted end-to-end)
• AI summaries use Claude API (opt-in, content is ephemeral)
• Voice transcription via Whisper API (opt-in, audio not stored by provider)
• No analytics, tracking, or third-party data sharing
• Export your entire archive anytime (open formats: JSON, Markdown, HTML)

Notes

• Archive database is SQLite (fully portable)
• Full-text search works offline
• Screenshots use headless Chrome (requires Chromium installed)
• Voice transcription requires OpenAI API key or local Whisper model
• Supports all major file types (PDF, images, videos, code, documents)
• Archive is yours forever — no vendor lock-in

---

Built for people who read things, save them, then never find them again. Not anymore. 📁🐾