Library Ebooks

My Amazon (Kindle) Library Ebook Lookup & Download

Overview

Internal tool to search and download ebooks from my personal Amazon (Kindle) library, i.e., titles I’ve purchased/added to my account (plus any personal documents synced into my library), across available formats.

Note: This skill is meant for my own licensed content. It does not bypass DRM. If a title can’t be downloaded in an accessible file format, use Amazon’s official reading/export options instead.

Prerequisites

IMPORTANT: Downloads require an Amazon-library access key for my account.

Before using download functionality, the user must:

1. Have access to their Amazon (Kindle) library (purchased titles / content library)
2. Set their API key/token: export LIBRARY_KEY="your-key"

The key is found in your internal library-proxy/account settings (or whatever system you use to authenticate access to your Amazon library index). Search works without a key, but downloads will fail.

If key is not set: Inform the user they need to set LIBRARY_KEY and provide the setup instructions above.

When to Use

• User asks to find/download a book from their Amazon library
• Need to look up content from a published book they own
• Searching for a specific edition or format that exists in their library
• "Get me the PDF of Clean Code (from my Kindle library)"
• "Find the latest edition of Design Patterns (that I own)"

Quick Reference

Environment Setup

export LIBRARY_KEY="your-membership-key"

The key is found in your Amazon library access setup (internal proxy/token settings).

Workflow

digraph download_flow {
    rankdir=TB;
    node [shape=box];

    search [label="Search by title/author"];
    verify [label="Verify correct book\n(check title, author, year)"];
    multiple [label="Multiple editions?" shape=diamond];
    prefer_recent [label="Prefer most recent\nunless specific edition requested"];
    format_ok [label="Preferred format available?" shape=diamond];
    download [label="Download via fast API"];
    convert [label="Use ebook-extractor\nto convert to text"];

    search -> verify;
    verify -> multiple;
    multiple -> prefer_recent [label="yes"];
    multiple -> format_ok [label="no"];
    prefer_recent -> format_ok;
    format_ok -> download [label="yes"];
    format_ok -> search [label="no - try different format"];
    download -> convert;
}

Common Patterns

Find and download a book

# Search with format preference
python3 scripts/library.py search "Clean Code Robert Martin" --format pdf --limit 5

# Verify it's the right book, get details
python3 scripts/library.py details adb5293cf369256a883718e71d3771c3

# Download
python3 scripts/library.py download adb5293cf369256a883718e71d3771c3 --output ./books/

Handle multiple editions

When search returns multiple editions:

1. Check year - prefer most recent unless user specified edition
2. Check format - match user's preference (pdf/epub)
3. Verify author matches exactly

Format Priority

Default priority when user doesn't specify: pdf > epub > mobi > azw3 > djvu

API Details

These endpoints refer to your internal Amazon-library index / proxy service (backed by your own Amazon account/library data), not a public “find-any-book” catalog.

Search endpoint: https://library-archive.org/search

• q - query string
• ext - format filter (pdf, epub, mobi, azw3, djvu)
• sort - year_desc for most recent first

Fast download API: https://library-archive.org/dyn/api/fast_download.json

• md5 - book identifier
• key - from LIBRARY_KEY env var

Common Mistakes

Converting to Text

Downloaded files are in their original format (PDF, EPUB, MOBI, etc.). To convert to plain text for analysis or processing, use the ebook-extractor skill after downloading.

Typical workflow:

1. Download with this skill → books/Clean_Code.pdf
2. Convert with ebook-extractor → books/Clean_Code.txt

Mirror Fallback

The script automatically tries multiple mirror domains if the primary domain is unavailable:

• library-archive.org (primary)
• library-archive.li
• library-archive.se
• library-archive.in
• library-archive.pm

The first working mirror is cached for the session. You'll see Using mirror: <domain> in stderr when a fallback is used.

Error Handling

• "Invalid md5" - MD5 hash is malformed or doesn't exist
• "Not a member" - Key is invalid or expired (i.e., your Amazon-library access token isn’t valid)
• No results - Broaden search terms, try author-only search
• "Could not connect to any mirror" - All mirrors are down, try again later

Troubleshooting

SSL Certificate Error on macOS

If you see this error:

[SSL: CERTIFICATE_VERIFY_FAILED] certificate verify failed: unable to get local issuer certificate

This happens because Python can't find the system's CA certificate bundle on macOS.

Quick Fix:

1. Install certifi:

   pip3 install certifi
   

2. Find your certificate path:

   python3 -c "import certifi; print(certifi.where())"
   

3. Add to ~/.zshrc:

   export SSL_CERT_FILE=/path/from/step/2/cacert.pem
   

4. Reload shell: source ~/.zshrc

Verify it works:

python3 -c "import urllib.request; urllib.request.urlopen('https://google.com')"

Why this happens: macOS uses Keychain for certificates, but Python doesn't use it by default. Framework installs (like /Library/Frameworks/Python.framework) often lack certificate configuration.

Do NOT use verify=False or PYTHONHTTPSVERIFY=0 - this disables SSL entirely and is insecure.