paperzilla
k-dense-ai/scientific-agent-skills
Access Paperzilla projects, recommendations, and canonical papers directly from your agent.
What is paperzilla?
Paperzilla is a research paper recommendation and curation platform. Use this skill when users ask for project recommendations, paper details, markdown summaries, feedback on recommendations, or feed exports.
- List and browse projects in Paperzilla
- Fetch and display canonical papers as markdown or JSON
- Open recommendations from projects and explain their relevance
- Leave feedback (upvote, downvote, star) on recommendations
- Export feeds as JSON or Atom URLs for feed readers
- Filter feeds by must-read status, date range, and limit
How to install paperzilla
npx skills add https://github.com/k-dense-ai/scientific-agent-skills --skill paperzilla- Install the `pz` CLI (via Homebrew, Scoop, or from source)
- Run `pz login` to authenticate with your Paperzilla account
- Have access to at least one Paperzilla project
How to use paperzilla
- 1.Install the pz CLI using the appropriate method for your OS
- 2.Run `pz login` to authenticate
- 3.List available projects with `pz project list`
- 4.Query a project's feed with `pz feed <project-id>` (add `--json` or `--atom` for export)
- 5.Fetch a paper with `pz paper <paper-id> --markdown` or view a recommendation with `pz rec <project-paper-id>`
- 6.Leave feedback on recommendations using `pz feedback <project-paper-id> upvote|downvote|star`
Use cases
- Get the latest paper recommendations for a specific research project
- Retrieve a canonical paper and generate a markdown summary for your team
- Export a project's feed to a feed reader using the Atom URL
- Leave feedback on whether a recommendation is relevant to your work
- Browse must-read papers from a project published since a specific date
- Researchers managing multiple paper projects
- Teams curating canonical papers for a domain
- Agents that need to query research recommendations programmatically
- Users integrating Paperzilla feeds into external tools
paperzilla FAQ
`pz paper` retrieves a canonical paper by ID, while `pz rec` opens a recommendation from one of your projects, which may include project-specific context and feedback markers.
Use `pz feed <project-id> --atom` to get a personal Atom feed URL that you can subscribe to in any feed reader.
`pz paper --markdown` only returns markdown if it is already prepared. For recommendations, `pz rec --markdown` can queue generation and will print a retry message while processing.
Use flags like `--since YYYY-MM-DD`, `--must-read`, and `--limit N` with the `pz feed` command.
You can upvote, downvote (with optional reason), or star a recommendation, and clear feedback with `pz feedback clear`.
Full instructions (SKILL.md)
Source of truth, from k-dense-ai/scientific-agent-skills.
name: paperzilla description: Chat with your agent about projects, recommendations, and canonical papers in Paperzilla. Use when users ask for recent project recommendations, canonical paper details, markdown-based summaries, recommendation feedback, feed export, or Atom feed URLs. license: MIT metadata: version: "1.0" skill-author: Paperzilla Inc
Paperzilla
Use this skill when you want to chat with your agent about projects, recommendations, and canonical papers in Paperzilla.
What you can ask
- "Give me the latest recommendations from project X."
- "Open recommendation Y and explain why it matters."
- "Fetch canonical paper Z as markdown and summarize it."
- "Tell me how this paper is relevant to my research."
- "Show me the feed for project X."
- "Leave feedback on a recommendation."
- "Export this paper, recommendation, or feed as JSON."
This is the core Paperzilla skill. It gives your agent direct access to Paperzilla data, but it does not impose a workflow or external delivery integration.
Access method
Most current profiles in this repo use the pz CLI.
If the current profile ships extra agent-specific instructions, follow those as well.
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
Use the official Linux install guide:
Build from source (Go 1.23+)
See the CLI repository for source builds:
Update
Check whether your CLI is up to date and get install-specific upgrade steps:
pz update
If detection is ambiguous, override it explicitly:
pz update --install-method homebrew
pz update --install-method scoop
pz update --install-method release
pz update --install-method source
Supported values are auto, homebrew, scoop, release, and source.
Authentication
pz login
CLI reference
If the current profile uses pz, these are the core commands.
List projects
pz project list
Show one project
pz project <project-id>
Browse project feed
pz feed <project-id>
Useful flags:
--must-read--since YYYY-MM-DD--limit N--json--atom
Examples:
pz feed <project-id> --must-read --since 2026-03-01 --limit 5
pz feed <project-id> --json
pz feed <project-id> --atom
Feed output can include existing recommendation feedback markers:
[↑]upvote[↓]downvote[★]star
Read a canonical paper
pz paper <paper-id>
pz paper <paper-id> --json
pz paper <paper-id> --markdown
pz paper <paper-id> --project <project-id>
Open a recommendation from one of your projects
pz rec <project-paper-id>
pz rec <project-paper-id> --json
pz rec <project-paper-id> --markdown
Leave recommendation feedback
pz feedback <project-paper-id> upvote
pz feedback <project-paper-id> star
pz feedback <project-paper-id> downvote --reason not_relevant
pz feedback clear <project-paper-id>
Output and automation
- Prefer
--jsonfor machine parsing. pz paper --markdownonly returns markdown when it is already prepared.pz rec --markdowncan queue markdown generation and prints a friendly retry message while it is still being prepared.--atomreturns a personal feed URL for feed readers.
Configuration
export PZ_API_URL="https://paperzilla.ai"
References
Related skills
More from k-dense-ai/scientific-agent-skills and the wider catalog.

parallel-web
Web search, data enrichment, and monitoring via Parallel CLI for current evidence and exhaustive research.

pathml
Local computational pathology workflows: slide tiling, preprocessing, multiplex quantification, and bounded model inference for research.

peer-review
Prepare evidence-bounded, constructive peer-review drafts with structured manuscript assessments and claim–evidence checks.

pennylane
Hardware-agnostic quantum ML framework with automatic differentiation for training quantum circuits.

phylogenetics
Build and analyze phylogenetic trees from sequence alignments using MAFFT, IQ-TREE 2, and FastTree.

polars
Lightning-fast DataFrame library for Python ETL, analytics, and pandas migration with lazy evaluation and parallel execution.