Paperzilla CLI
Verified@pors
npx machina-cli add skill @pors/pz --openclawPaperzilla CLI (pz) 🦖
A command-line tool for Paperzilla, an AI-powered scientific paper discovery platform.
Install
macOS:
brew install paperzilla-ai/tap/pz
Windows (Scoop):
scoop bucket add paperzilla-ai https://github.com/paperzilla-ai/scoop-bucket
scoop install pz
Linux:
curl -sL https://github.com/paperzilla-ai/pz/releases/latest/download/pz_linux_amd64.tar.gz | tar xz
sudo mv pz /usr/local/bin/
Authentication
Log in with your Paperzilla account before doing anything else:
pz login
Core Commands
List Projects
Lists your available projects and their IDs.
pz project list
Browse Feed
Fetches the papers for a given project ID.
pz feed <project-id>
Filter & Export Flags:
--must-read: Filter only for must-read papers.--limit 5: Limit the number of returned results.--since YYYY-MM-DD: Filter papers published after a specific date.--json: Output in JSON format (excellent for parsing with other tools or piping into Claude/Gemini).
Example:
pz feed 12345 --must-read --limit 5 --json
Feed Reader Integration (RSS/Atom)
Get an Atom feed URL you can add to any feed reader (Vienna RSS, NetNewsWire, Feedly, etc.):
pz feed <project-id> --atom
This prints a URL with an embedded feed token. Running --atom again after revoking will generate a new token.
API Configuration
By default, it connects to https://paperzilla.ai. To override, set PZ_API_URL.
export PZ_API_URL="https://custom.paperzilla.ai"
Overview
The Paperzilla CLI (pz) lets you search, filter, and browse AI-curated academic papers from Paperzilla directly in your terminal. It supports listing your projects, fetching feeds by project ID, and exporting results, with optional JSON output or RSS/Atom feeds. Authentication via a Paperzilla account is required to access your data.
How This Skill Works
Install the CLI via brew, scoop, or a direct download, then run pz login to authenticate. Use pz project list to see your projects and pz feed <project-id> to retrieve papers, applying filters like --must-read, --limit, --since, and --json for machine-friendly output or downstream tooling; add --atom to obtain an RSS/Atom feed URL for readers.
When to Use It
- When you want to see all projects you have in Paperzilla and their IDs
- When you need to fetch and review papers from a specific project feed
- When you want to filter results to must-read papers or limit results for quick review
- When you’re automating workflows and need JSON output for pipelines
- When you want an RSS/Atom feed URL to read papers in a feed reader
Quick Start
- Step 1: Install Paperzilla CLI (brew, scoop, or curl-based install)
- Step 2: Run pz login to authenticate with your Paperzilla account
- Step 3: Browse or fetch papers, e.g., pz project list or pz feed <project-id> --json
Best Practices
- Always run pz login before accessing data to ensure authenticated requests
- Use --json for easy integration with other tools or scripts
- Combine --limit with --since to manage result scope and recency
- Use --atom to generate an RSS/Atom feed URL for your reader
- Set PZ_API_URL when using a non-default Paperzilla endpoint for testing or staging
Example Use Cases
- pz project list to enumerate available projects and IDs
- pz feed 12345 --limit 5 --json to fetch the latest five papers in JSON
- pz feed 12345 --must-read --limit 10 --json | jq .
- pz feed 12345 --atom to obtain an RSS/Atom feed URL for a reader
- export PZ_API_URL="https://custom.paperzilla.ai" && pz login && pz feed 12345 --json
