paperzilla

Use the Paperzilla CLI (pz) to search, filter, and browse high-signal academic papers. Trigger when the user wants to check research feeds, list projects, inspect must-read papers, export JSON results, or generate Atom feed links from Paperzilla. Requires a Paperzilla account and pz login.

# Paperzilla CLI (pz) 🦖

A command-line tool for [Paperzilla](https://paperzilla.ai), an AI-powered scientific paper discovery platform.

## Install

macOS:
```bash
brew install paperzilla-ai/tap/pz
```

Windows (Scoop):
```bash
scoop bucket add paperzilla-ai https://github.com/paperzilla-ai/scoop-bucket
scoop install pz
```

Linux:
```bash
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:
```bash
pz login
```

## Core Commands

### List Projects
Lists your available projects and their IDs.
```bash
pz project list
```

### Browse Feed
Fetches the papers for a given project ID.
```bash
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:
```bash
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.):
```bash
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`.
```bash
export PZ_API_URL="https://custom.paperzilla.ai"
```