Reads projects, searches project feeds, and retrieves recommendations and canonical papers in Paperzilla through the pz CLI.

MITAuto-check passedResearch & Science

Install Paperzilla

skills CLI
$ npx skills add K-Dense-AI/scientific-agent-skills --skill paperzilla -a claude-code

Project install by default; add -g for ~/.claude/skills/.

GitHub CLI
$ gh skill install K-Dense-AI/scientific-agent-skills paperzilla --agent claude-code

Project scope by default; add --scope user for a personal install. Needs GitHub CLI 2.90.0 or later (public preview).

Manual copy
$ git clone --depth 1 https://github.com/K-Dense-AI/scientific-agent-skills.git skills-src && mkdir -p .claude/skills && cp -r skills-src/skills/paperzilla .claude/skills/paperzilla && rm -rf skills-src

Use ~/.claude/skills/ instead of .claude/skills for a personal install. The folder must contain SKILL.md.

Claude Code skills documentation · loads skills from .claude/skills/

Facts

Skill name
paperzilla
GitHub stars
48k
Used in
1 other repo
Token cost
~2.3k tokens
SKILL.md length
999 words
Files
2 (incl. references)
Skills in repo
153
Repo updated
First seen
Licence
MIT

At a glance

Reads projects, searches project feeds, and retrieves recommendations and canonical papers in Paperzilla through the pz CLI.

  • Research & Science work in your project
  • SKILL.md covers What you can ask, Access method, Install and Update, plus 6 more sections
  • Calls brew; reaches github.com and paperzilla.ai

What it does

Paperzilla is an agent skill from K-Dense-AI/scientific-agent-skills. Reads projects, searches project feeds, and retrieves recommendations and canonical papers in Paperzilla through the pz CLI. Supports recent recommendations, paper details, markdown-based summaries, recommendation feedback, JSON export, and Atom feed URLs.

Its SKILL.md is about 2.3k tokens, which your agent loads only when the skill is triggered. The skill folder holds 2 other files, including reference files (for example `references/cli-contracts.md`). Compatibility notes: Requires the pz CLI, network access, and a Paperzilla account with CLI access. Source builds require Go 1.23 or later.

It sits in Research & Science. The repository describes itself as: Turn any AI agent into an AI Scientist. The 1 Agent Skills library for science, used by 250,000+ scientists worldwide. 177 ready-to-use validated skills plus 100+ scientific… The licence is MIT.

When your agent uses it

  • Research & Science work in your project

Example prompts

  • “Use the paperzilla skill to read projects, searches project feeds, and retrieves recommendations and canonical papers in Paperzilla through the pz CLI”
  • “/paperzilla”

Requirements

  • Compatibility (from SKILL.md): Requires the pz CLI, network access, and a Paperzilla account with CLI access. Source builds require Go 1.23 or later.

What it can do on your machine

Read from SKILL.md and the folder at commit 92ace75. It shows what the files ask for, not the result of running them.

  • Tool permissions

    Pre-approves nothing: there is no allowed-tools line, so your agent's usual permission prompts apply.

    From allowed-tools in the SKILL.md frontmatter.

  • Runs code

    Shell commands in SKILL.md call:

    • brew

    From the folder's file list and the shell code blocks in SKILL.md.

  • Network

    Hosts in commands or code, which the agent is likely to contact:

    • github.com
    • paperzilla.ai

    Also links to:

    • docs.paperzilla.ai

    From URLs in SKILL.md, links to its own repository left out.

  • Credentials

    Names no API keys, tokens, secrets or passwords.

    From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.

  • Compatibility

    Requires the pz CLI, network access, and a Paperzilla account with CLI access. Source builds require Go 1.23 or later.

    From compatibility in the SKILL.md frontmatter.

Context cost

Paperzilla loads about 2.3k tokens when it runs, and up to ~4.2k if it reads all its reference files. Until then it costs about 67 tokens; SKILL.md has 999 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it
~67
When it runs · the whole SKILL.md, loaded when a task matches
~2.3k
With references · SKILL.md plus every file in references/, read only if the agent opens them
~4.2k

Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.

Safety

Auto-check passed

The automated check found no risky patterns in SKILL.md.

Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); files beside SKILL.md are not scanned.

SKILL.md

The full file from K-Dense-AI/scientific-agent-skills at commit 92ace75, republished under its MIT licence (© K-Dense-AI). 999 words, ~2,324 tokens.

Download SKILL.mdSave it as .claude/skills/paperzilla/SKILL.md (or your agent's skills folder). This skill also uses 1 other file; get the full folder from GitHub.
name
paperzilla
description
Reads projects, searches project feeds, and retrieves recommendations and canonical papers in Paperzilla through the pz CLI. Supports recent recommendations, paper details, markdown-based summaries, recommendation feedback, JSON export, and Atom feed URLs.
compatibility
Requires the pz CLI, network access, and a Paperzilla account with CLI access. Source builds require Go 1.23 or later.
license
MIT
metadata.version
1.2
metadata.skill-author
Paperzilla Inc
metadata.upstream-version
pz 0.7.1
metadata.last-reviewed
2026-10-01

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

Use the official pz CLI. This skill targets 0.7.1, checked against the official release source and binary. Command examples use placeholder IDs; replace them with records returned by your account. CLI behavior was tested against a local mock server, not an authenticated Paperzilla account.

Start with pz --version, then list projects, select a project, browse or search its feed, and inspect selected recommendations. Keep project recommendations separate from the underlying canonical paper when summarizing or exporting.

Install

macOS
bash
brew install paperzilla-ai/tap/pz

If Homebrew 6 or later rejects this formula as untrusted, the official guide uses brew trust --formula paperzilla-ai/tap/pz, then retries installation. Trust only that formula. Installation and upgrade commands were documentation-checked, not executed during this review.

Windows (Scoop)
bash
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:

bash
pz update

If detection is ambiguous, override it explicitly:

bash
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

bash
pz login

Login sends a one-time code to the account email and saves the session locally. Complete it interactively before running JSON exports: missing or expired credentials can trigger login prompts on stdout. The CLI checks account access and refreshes tokens automatically; report access/upgrade errors rather than treating them as an empty feed.

Version-specific discrepancy: the online guides describe anonymous canonical paper access, but the 0.7.1 binary requires login and a CLI-access check even for pz paper. Its subsequent canonical-paper HTTP request uses a public route. Do not promise an anonymous CLI workflow for this release.

CLI reference

If the current profile uses pz, these are the core commands.

List projects
bash
pz project list
pz project list --json
Show one project
bash
pz project <project-id>
pz project <project-id> --json

The list JSON is an array containing id, name, mode, and visibility. The single-project JSON also includes keywords, watched sources, and categories.

Browse project feed
bash
pz feed <project-id>

Useful flags:

  • --must-read
  • --since YYYY-MM-DD
  • --limit N
  • --offset N (zero-based results to skip)
  • --json
  • --atom

Examples:

bash
pz feed <project-id> --must-read --since 2026-03-01 --limit 5
pz feed <project-id> --limit 20 --offset 20 --json
pz feed <project-id> --json
pz feed <project-id> --atom

--since filters when recommendations were ready, not paper publication date. Browse JSON contains items, total, limit, and offset. A command fetches one page; it does not export the entire feed automatically. Retain the same filters, advance from the returned offset by the number of items received, and stop on an empty page or when the returned total is reached. Deduplicate by recommendation ID if the live feed changes while paging.

Feed output can include existing recommendation feedback markers:

  • [↑] upvote
  • [↓] downvote
  • [★] star
Search the full project feed
bash
pz feed search --project-id <project-id> --query "latent retrieval" --json
pz feed search --project-id <project-id> --query "retrieval" --feedback-filter starred --must-read --limit 20 --offset 20 --json

Search ranks results across the whole project feed. Use a trimmed query of 3–200 characters and a positive --limit up to 100. --offset must be nonnegative. Feedback filters are all, unrated, liked, disliked, starred, not-relevant, and low-quality; these hyphenated filters differ from the underscored downvote reasons below. Search does not accept --since.

Search JSON contains items, limit, offset, has_more, and query, with no exact total. Request the next offset only while has_more is true, keeping the query and filters unchanged. Stop and report an inconsistent empty page rather than looping indefinitely. The second example illustrates a subsequent page.

Show full SKILL.md (387 more words)Show less
Read a canonical paper
bash
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
bash
pz rec <project-paper-id>
pz rec <project-paper-id> --json
pz rec <project-paper-id> --markdown
Leave recommendation feedback
bash
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>

Feedback changes account data; use it when the user requests that change. Downvote reasons are not_relevant or low_quality, and --reason applies only to a downvote. Add --json to get the feedback object; clearing returns {"project_paper_ref": "...", "cleared": true}. clear is a subcommand before the recommendation ID.

Keep paper and recommendation identities separate

A canonical paper-id identifies the paper; a project-paper-id identifies its recommendation within a project. Take both from returned records and retain the project association when exporting results. Use the recommendation ID for rec and feedback operations, even if the same paper appears in several projects. In feed JSON, items[].id/items[].short_id are recommendation identifiers; items[].paper.id/items[].paper.short_id are canonical paper identifiers. Do not infer recommendation IDs from a DOI or canonical paper ID. See the official CLI documentation.

When markdown is still being prepared, report that state and summarize only the metadata or abstract actually returned. A retry message is not full-text evidence.

Output and automation

  • Prefer --json for machine parsing after completing login. paper and rec reject combining --json with --markdown.
  • pz paper --markdown only returns markdown when it is already prepared.
  • pz rec --markdown and project-scoped pz paper --project ... --markdown can queue markdown generation. A pending message can exit successfully; check the content before treating stdout as full text. Avoid repeated immediate retries and report the pending state.
  • --atom returns a personal feed URL for feed readers and can create a feed token. Anyone holding that URL can read the feed; keep its token out of logs, public reports, and repositories. It does not return Atom XML or page JSON, and browse filters are not applied to the generated URL.
  • A relevance score expresses project matching, not scientific validity or an effect estimate. Label whether a summary used metadata, an abstract, or actual markdown. Missing PDF URLs do not imply that a source landing page is absent.

Configuration

bash
export PZ_API_URL="https://paperzilla.ai"

The default already points to Paperzilla; change it only for a trusted service or a local test server. Authenticated requests send session credentials there. PZ_TOKENS_PATH optionally selects a session file; by default the CLI uses ~/.paperzilla/tokens.json. Never export that file as research data.

See the verified CLI contracts for request routes, response shapes, pagination, and the documentation/source discrepancy.

References

© K-Dense-AI, MIT. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file

Files

SKILL.md and 1 other file (references) in skills/paperzilla of K-Dense-AI/scientific-agent-skills.

  • SKILL.md
  • references/cli-contracts.md

Open the folder on GitHubat commit 92ace75

Used in 1 other repository

We found 1 copy of this SKILL.md (exact, near-identical or edited) in other folders, from 1 other GitHub owner. This page covers the copy in K-Dense-AI/scientific-agent-skills, which our catalogue first saw on October 7, 2026.

Compare with similar skills

Paperzilla next to the 5 skills that share the most tags, products or categories with it. Stars are the repository's; “used in” counts other GitHub owners with a copy.

Paperzilla compared with similar skills
SkillStarsUsed inTokensAuto-checkLicenceRepo updated
Paperzilla this skillK-Dense-AI/scientific-agent-skills48k1 repos~2.3kAutomated safety check: PassMIT
Hypothesis Generationspacering-net/codeg3.9k14 repos~3.6kAutomated safety check: NotesMIT
GitHub Deep Researchbytedance/deer-flow84k4 repos~1.3kAutomated safety check: PassMIT
Nature Paper CardYuan1z0825/nature-skills47k2 repos~2.1kAutomated safety check: PassApache-2.0
Content Research Writerweapp-tailwindcss/weapp-tailwindcss1.9k25 repos~3.5kAutomated safety check: PassMIT
Peer Reviewspacering-net/codeg3.9k17 repos~5.9kAutomated safety check: NotesMIT

Similar skills

  • Hypothesis Generation

    spacering-net/codeg

    Structured hypothesis formulation from observations. An agent skill from spacering-net/codeg.

    3.9k GitHub starsUsed in 14 repos~3.6k tokens
    Research & ScienceAuto-check: notes
  • GitHub Deep Research

    bytedance/deer-flow

    Researches a GitHub repository over four rounds using the GitHub API and web search, then writes a structured markdown report with timeline, metrics and Mermaid diagrams.

    84k GitHub starsUsed in 4 repos~1.3k tokens
    Research & ScienceAuto-check passed
  • Nature Paper Card

    Yuan1z0825/nature-skills

    Builds a structured deep-reading card for one scientific paper, covering methods, how experiments support claims, limitations and research ideas, with a script to prepare the source.

    47k GitHub starsUsed in 2 repos~2.1k tokens
    Research & ScienceAuto-check passed
  • Content Research Writer

    weapp-tailwindcss/weapp-tailwindcss

    Assists in writing high-quality content by conducting research, adding citations, improving hooks, iterating on outlines, and providing real-time feedback on each section.

    1.9k GitHub starsUsed in 25 repos~3.5k tokens
    Research & ScienceAuto-check passed
  • Peer Review

    spacering-net/codeg

    Structured manuscript/grant review with checklist-based evaluation.

    3.9k GitHub starsUsed in 17 repos~5.9k tokens
    Research & ScienceAuto-check: notes
  • Last30days

    mvanhorn/last30days-skill

    Research what people actually say about any topic in the last 30 days.

    64k GitHub stars~7.9k tokensUpdated today
    Research & ScienceAuto-check: notes

More from K-Dense-AI/scientific-agent-skills

All 153 skills in this repo
  • 13C Metabolic Flux Analysis

    K-Dense-AI/scientific-agent-skills

    Estimates reaction fluxes inside cells from steady-state carbon-13 labeling data with a bundled mfapy-based solver, and reports which fluxes the data pin down.

    48k GitHub starsUsed in 1 repo~3.2k tokens
    Auto-check passed
  • Analytical Method Validation Planner

    K-Dense-AI/scientific-agent-skills

    Plans, runs, and documents analytical method validation, verification, or transfer studies under ICH Q2(R2)/Q14, USP, ICH M10, CLSI EP, or ISO/IEC 17025.

    48k GitHub starsUsed in 1 repo~4.9k tokens
    Auto-check: notes
  • Cantera Ignition Delay

    K-Dense-AI/scientific-agent-skills

    Runs Cantera constant-volume or constant-pressure ignition simulations and reports temperature-based ignition delay with mechanism provenance and checks.

    48k GitHub starsUsed in 1 repo~2.2k tokens
    Auto-check passed
  • DiffDock Molecular Docking

    K-Dense-AI/scientific-agent-skills

    Predicts how small molecules bind to a protein with DiffDock, covering batch docking, pose ranking by confidence and checks on the results; not for binding affinity.

    48k GitHub starsUsed in 1 repo~3k tokens
    Auto-check: notes
  • HypoGeniC Hypothesis Generation

    K-Dense-AI/scientific-agent-skills

    Plans and audits runs of the HypoGeniC and HypoRefine packages, which propose hypotheses from labeled text datasets, with local checks before any model call.

    48k GitHub starsUsed in 1 repo~3.6k tokens
    Auto-check: notes
  • ISO Standards Readiness Evidence

    K-Dense-AI/scientific-agent-skills

    Organizes scope, controlled documents, risk files and traceability into draft evidence for human review against ISO 13485, 14971, 17025 and 15189.

    48k GitHub starsUsed in 1 repo~4.6k tokens
    Auto-check: notes

Questions about Paperzilla

What does Paperzilla do?

Reads projects, searches project feeds, and retrieves recommendations and canonical papers in Paperzilla through the pz CLI. Paperzilla is an agent skill from K-Dense-AI/scientific-agent-skills. Reads projects, searches project feeds, and retrieves recommendations and canonical papers in Paperzilla through the pz CLI.

When should I use Paperzilla?

Paperzilla fits situations like: research & Science work in your project.

How do I install Paperzilla in Claude Code?

Run `npx skills add K-Dense-AI/scientific-agent-skills --skill paperzilla -a claude-code`. Or copy the skill folder (skills/paperzilla in K-Dense-AI/scientific-agent-skills) into .claude/skills/paperzilla in your project. Claude Code loads it when a task matches its description.

How do I install Paperzilla in Codex?

Run `npx skills add K-Dense-AI/scientific-agent-skills --skill paperzilla -a codex`. Or copy the skill folder (skills/paperzilla in K-Dense-AI/scientific-agent-skills) into .agents/skills/paperzilla in your project. Codex loads it when a task matches its description.

Can I use Paperzilla in Cursor, Gemini CLI or GitHub Copilot?

Cursor, Gemini CLI, GitHub Copilot and OpenCode also load SKILL.md folders. With the skills CLI, run `npx skills add K-Dense-AI/scientific-agent-skills --skill paperzilla -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/paperzilla, .gemini/skills/paperzilla, .github/skills/paperzilla and .opencode/skills/paperzilla in your project.

What does Paperzilla need to run?

Going by SKILL.md and its folder, Paperzilla needs the command-line tools its instructions call (brew). Compatibility (from SKILL.md): Requires the pz CLI, network access, and a Paperzilla account with CLI access. Source builds require Go 1.23 or later..

Does Paperzilla access the network?

SKILL.md names 3 domains. In commands or code: github.com and paperzilla.ai; the agent is likely to contact these when it follows the instructions. As links in the text: docs.paperzilla.ai. This is read from the text; nothing was executed.

Is Paperzilla safe to install?

Our automated static check of SKILL.md found no risky patterns, such as piping downloads into a shell, reading credential files or hidden Unicode. It is not a guarantee. Review the folder before installing.

What licence does Paperzilla use?

Paperzilla is published under the MIT licence (declared in SKILL.md). It allows redistribution, so the full SKILL.md is shown on this page.

How many tokens does Paperzilla use?

About 2.3k tokens (SKILL.md is roughly 9.3k characters). Agents keep only the skill's name and description in context until a task matches; then they load SKILL.md in full. Its references folder adds about 1.8k tokens, read only when the agent opens those files.

What are the alternatives to Paperzilla?

Skills that share tags, products or a category with Paperzilla: Hypothesis Generation (spacering-net/codeg, 3.9k stars), GitHub Deep Research (bytedance/deer-flow, 84k stars), Nature Paper Card (Yuan1z0825/nature-skills, 47k stars) and Content Research Writer (weapp-tailwindcss/weapp-tailwindcss, 1.9k stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.

Who maintains Paperzilla?

K-Dense-AI (a GitHub organization) maintains it in K-Dense-AI/scientific-agent-skills, which has 48,095 GitHub stars. The repository holds 153 skills in this directory. The repository was last updated on October 5, 2026.

Source: K-Dense-AI/scientific-agent-skills on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.