Agent skill

Search

Unified academic paper search, citation chains, paper download (arXiv LaTeX/PDF, Sci-Hub), figure extraction from papers, LaTeX source reading, BibTeX fetching, web search, and browser automation…

MITAuto-check: warningsResearch & Science

Install Search

The automated check flagged lines worth reading first. See the safety section below.

skills CLI

$ npx skills add Muuuun/luxas --skill search -a claude-code

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

GitHub CLI

$ gh skill install Muuuun/luxas search --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/Muuuun/luxas.git skills-src && mkdir -p .claude/skills && cp -r skills-src/skills/search .claude/skills/search && 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/

Paste into Claude Code

Install the "search" agent skill from https://github.com/Muuuun/luxas/tree/main/skills/search into .claude/skills/search/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "search", then confirm the skill loads.

Claude Code copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.

$skill-installer

$skill-installer install https://github.com/Muuuun/luxas/tree/main/skills/search

Type this inside Codex. $skill-installer <name> installs a curated skill from openai/skills. The installer writes to $CODEX_HOME/skills (default ~/.codex/skills). Restart Codex if the skill does not show up.

skills CLI

$ npx skills add Muuuun/luxas --skill search -a codex

Project install goes to .agents/skills/; add -g for ~/.codex/skills/.

GitHub CLI

$ gh skill install Muuuun/luxas search --agent codex

Project scope by default (.agents/skills/); add --scope user for a personal install.

Manual copy

$ git clone --depth 1 https://github.com/Muuuun/luxas.git skills-src && mkdir -p .agents/skills && cp -r skills-src/skills/search .agents/skills/search && rm -rf skills-src

Use ~/.agents/skills/ instead of .agents/skills for a personal install.

Codex skills documentation · loads skills from .agents/skills/

Paste into Codex

Install the "search" agent skill from https://github.com/Muuuun/luxas/tree/main/skills/search into .agents/skills/search/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "search", then confirm the skill loads.

Codex copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.

skills CLI

$ npx skills add Muuuun/luxas --skill search -a cursor

Project install goes to .agents/skills/; add -g for ~/.cursor/skills/.

GitHub CLI

$ gh skill install Muuuun/luxas search --agent cursor

Project scope by default (.agents/skills/); add --scope user for a personal install.

Manual copy

$ git clone --depth 1 https://github.com/Muuuun/luxas.git skills-src && mkdir -p .cursor/skills && cp -r skills-src/skills/search .cursor/skills/search && rm -rf skills-src

Use ~/.cursor/skills/ instead of .cursor/skills for a personal install.

Cursor skills documentation · loads skills from .cursor/skills/, .agents/skills/, .claude/skills/, .codex/skills/

Paste into Cursor

Install the "search" agent skill from https://github.com/Muuuun/luxas/tree/main/skills/search into .cursor/skills/search/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "search", then confirm the skill loads.

Cursor copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.

gemini skills install

$ gemini skills install https://github.com/Muuuun/luxas.git --path skills/search

--scope user (default) or --scope workspace; --path is the subfolder of the repo that holds the skill; --consent skips the security confirmation prompt.

skills CLI

$ npx skills add Muuuun/luxas --skill search -a gemini-cli

Project install goes to .agents/skills/; add -g for ~/.gemini/skills/.

GitHub CLI

$ gh skill install Muuuun/luxas search --agent gemini-cli

Project scope by default (.agents/skills/); add --scope user for a personal install.

Manual copy

$ git clone --depth 1 https://github.com/Muuuun/luxas.git skills-src && mkdir -p .gemini/skills && cp -r skills-src/skills/search .gemini/skills/search && rm -rf skills-src

Use ~/.gemini/skills/ instead of .gemini/skills for a personal install, then run /skills reload.

Gemini CLI skills documentation · loads skills from .gemini/skills/, .agents/skills/

Paste into Gemini CLI

Install the "search" agent skill from https://github.com/Muuuun/luxas/tree/main/skills/search into .gemini/skills/search/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "search", then confirm the skill loads.

Gemini CLI copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.

GitHub CLI

$ gh skill install Muuuun/luxas search

Installs for Copilot at project scope by default; add --scope user for a personal install. Preview a skill first with gh skill preview. Needs GitHub CLI 2.90.0 or later (public preview).

skills CLI

$ npx skills add Muuuun/luxas --skill search -a github-copilot

Project install goes to .agents/skills/; add -g for ~/.copilot/skills/.

Manual copy

$ git clone --depth 1 https://github.com/Muuuun/luxas.git skills-src && mkdir -p .github/skills && cp -r skills-src/skills/search .github/skills/search && rm -rf skills-src

Use ~/.copilot/skills/ instead of .github/skills for a personal install. Commit .github/skills so cloud agent and code review can use it.

GitHub Copilot skills documentation · loads skills from .github/skills/, .claude/skills/, .agents/skills/

Paste into GitHub Copilot

Install the "search" agent skill from https://github.com/Muuuun/luxas/tree/main/skills/search into .github/skills/search/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "search", then confirm the skill loads.

GitHub Copilot copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.

skills CLI

$ npx skills add Muuuun/luxas --skill search -a opencode

OpenCode documents no install command of its own. Project install goes to .agents/skills/; add -g for ~/.config/opencode/skills/.

GitHub CLI

$ gh skill install Muuuun/luxas search --agent opencode

Project scope by default (.agents/skills/); add --scope user for a personal install.

Manual copy

$ git clone --depth 1 https://github.com/Muuuun/luxas.git skills-src && mkdir -p .opencode/skills && cp -r skills-src/skills/search .opencode/skills/search && rm -rf skills-src

Use ~/.config/opencode/skills/ instead of .opencode/skills for a personal install.

OpenCode skills documentation · loads skills from .opencode/skills/, .claude/skills/, .agents/skills/

Paste into OpenCode

Install the "search" agent skill from https://github.com/Muuuun/luxas/tree/main/skills/search into .opencode/skills/search/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "search", then confirm the skill loads.

OpenCode copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.

Facts

Skill name: search
GitHub stars: 1.2k
Token cost: ~2.3k tokens
SKILL.md length: 717 words
Files: 8 (incl. scripts)
Skills in repo: 9
Repo updated: Sep 6, 2026
First seen: Oct 7, 2026
Licence: MIT

At a glance

Unified academic paper search, citation chains, paper download (arXiv LaTeX/PDF, Sci-Hub), figure extraction from papers, LaTeX source reading, BibTeX fetching, web search, and browser automation…

Tasks that involve Academic paper search
SKILL.md covers Setup, scripts/search, scripts/browse and Decision Guide
Runs Shell scripts from its folder; calls bash, gh and git; reaches journals.aps.org; needs BRAVE_API_KEY and GITHUB_TOKEN
Tasks that involve Citation management

What it does

Search is an agent skill from Muuuun/luxas. Unified academic paper search, citation chains, paper download (arXiv LaTeX/PDF, Sci-Hub), figure extraction from papers, LaTeX source reading, BibTeX fetching, web search, and browser automation for Cloudflare-protected sites (PRL, Science, Nature, Google Scholar).

Its SKILL.md is about 2.3k tokens, which your agent loads only when the skill is triggered. The skill folder holds 8 other files, including scripts (for example `scripts/setup.sh`). Compatibility notes: Requires Node.js 22+. Optional: browser-use CLI for browser automation. PyMuPDF and Pillow for figure extraction.

It sits in Research & Science, covering Academic paper search, Citation management and LaTeX. It works with LaTeX, arXiv and Cloudflare. The repository describes itself as: An autonomous research colleague — from a question to a compiled manuscript, while you sleep. The licence is MIT.

When your agent uses it

Tasks that involve Academic paper search
Tasks that involve Citation management
Tasks that involve LaTeX

Example prompts

“/search”

Requirements

Node.js
A Bash shell
A credential in BRAVE_API_KEY
A credential in GITHUB_TOKEN
Compatibility (from SKILL.md): Requires Node.js 22+. Optional: browser-use CLI for browser automation. PyMuPDF and Pillow for figure extraction.
Pre-approved tools (allowed-tools): Bash(search:*), Bash(browse:*), Bash(browser-use:*), Bash(extract-figures:*)

What it can do on your machine

Read from SKILL.md and the folder at commit 9f77cef. It shows what the files ask for, not the result of running them.

Tool permissions
Pre-approves these tools, so the agent can use them without asking each time:
- Bash(search:*)
- Bash(browse:*)
- Bash(browser-use:*)
- Bash(extract-figures:*)
From allowed-tools in the SKILL.md frontmatter.
Runs code
Ships 7 files in scripts/ (Shell), which the agent can run.
Shell commands in SKILL.md call:
- bash
- gh
- git
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:
- journals.aps.org
From URLs in SKILL.md, links to its own repository left out.
Credentials
Names these keys or tokens, usually read from environment variables:
- BRAVE_API_KEY
- GITHUB_TOKEN
From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.
Compatibility
Requires Node.js 22+. Optional: browser-use CLI for browser automation. PyMuPDF and Pillow for figure extraction.
From compatibility in the SKILL.md frontmatter.

Context cost

Search loads about 2.3k tokens when it runs. Until then it costs about 68 tokens; SKILL.md has 717 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it: ~68
When it runs · the whole SKILL.md, loaded when a task matches: ~2.3k

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: warnings

The automated check found patterns that need a careful read before installing.

WarningMentions a credentials file (SSH keys, cloud or package-manager tokens)SKILL.md:129

hes in headed mode with the user's real Chrome profile (cookies, fingerprint), so it bypasses Cloudflare and Google Scho

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); the scripts in this folder are not scanned.

SKILL.md

The full file from Muuuun/luxas at commit 9f77cef, republished under its MIT licence (© Muuuun). 717 words, ~2,346 tokens.

Download SKILL.mdSave it as .claude/skills/search/SKILL.md (or your agent's skills folder). This skill also uses 7 other files; get the full folder from GitHub.

name: search
description: Unified academic paper search, citation chains, paper download (arXiv LaTeX/PDF, Sci-Hub), figure extraction from papers, LaTeX source reading, BibTeX fetching, web search, and browser automation for Cloudflare-protected sites (PRL, Science, Nature, Google Scholar).
allowed-tools: Bash(search:*), Bash(browse:*), Bash(browser-use:*), Bash(extract-figures:*)
compatibility: Requires Node.js 22+. Optional: browser-use CLI for browser automation. PyMuPDF and Pillow for figure extraction.

Search Skill

All information gathering in one place. Two CLI tools:

scripts/search — paper search, citations, download, LaTeX source, BibTeX, web search, URL fetch
scripts/browse — browser automation wrapper (uses browser-use with headed + Chrome profile defaults)

Setup

bash

bash scripts/setup.sh

scripts/search

Paper Search

bash

scripts/search papers "Rydberg atom quantum gate"
scripts/search papers "topological insulators" --source openalex --count 20
scripts/search papers "diffusion models" --source arxiv --count 15
scripts/search papers "air pollution health effects" --source crossref --count 20
scripts/search papers "quantum error correction" --from-year 2024 --sort date --count 20
# Author-filtered search: use last name as an indexed field, not free text.
# At least one of <query> or --author is required; both together = AND.
scripts/search papers "" --author "Lukin" --from-year 2025 --sort date --count 20
scripts/search papers "neutral atom arrays" --author "Bluvstein" --count 15

OpenAlex: all fields, citation counts, DOIs. --source openalex
arXiv: recent preprints, physics/CS/math. --source arxiv
CrossRef: broadest coverage (120M+ records), all disciplines, non-English journals. --source crossref
Default: all three.
--author <LastName>: Filter by author last name. Maps to arXiv au:, OpenAlex raw_author_name.search, and CrossRef query.author — all indexed fields. Strongly preferred over putting the name in the free-text query, which treats the name as an unweighted keyword and routinely misses the target. Pair with --from-year for recent work by a specific researcher.
--from-year YYYY: Only return papers published in YYYY or later.
--sort relevance|date: Sort by relevance (default) or publication date (newest first).

Citation Chains

bash

scripts/search citations W2057883617
scripts/search citations "10.1038/s41586-021-03819-2" --direction citations --limit 30
scripts/search citations 2301.07041 --direction references

Accepts: OpenAlex ID (W...), DOI, or arXiv ID. --direction: citations (forward), references (backward), both (default).

Paper Download

bash

scripts/search download --arxiv 2301.07041 --output data/papers
scripts/search download --doi "10.1038/s41586-021-03819-2" --output data/papers
scripts/search download --url "https://example.com/paper.pdf" --output data/papers

arXiv: tries LaTeX source first (tar.gz), falls back to PDF.
DOI: auto-discovers working Sci-Hub mirror, downloads PDF.
URL: direct download.
Default --output: data/papers/

Automatic reader dispatch (arXiv only). After a successful search download --arxiv <id>, a background reader is dispatched automatically. It reads the paper's main text + figure manifest and writes two things in a single pass: methodology coverage (A/B/C/D: what's computed, what's demo'd, what goes in figures, what rigor bar) into notes/methodology.md, AND a per-paper literature entry (keyed by cite_key) into notes/literature.md. You never need to launch the reader manually. If it silently misses a paper (hook race, manual drop, DOI download), the research snapshot will surface an <unprocessed_papers> reminder asking you to spawn_agent(agent="reader", ...) for the gap.

LaTeX Source

bash

scripts/search source 2301.07041
scripts/search source 2301.07041 --papers-dir data/papers --max-chars 50000

Downloads if not present, then returns concatenated .tex/.bib content. Much better than PDF for extracting equations, methods, citations.

BibTeX

bash

scripts/search bib "10.1038/s41586-021-03819-2"
scripts/search bib "10.1038/s41586-021-03819-2" --save report/references.bib

Fetches BibTeX via doi.org content negotiation, falls back to CrossRef. --save appends to a .bib file (deduplicates by DOI).

Web Search (Brave)

bash

scripts/search web "quantum error correction review 2024" --count 10

Requires BRAVE_API_KEY env var.

GitHub

bash

# Repo search — find repositories by keyword
scripts/search github "quantum simulation" --count 10
scripts/search github "transformer inference" --from-year 2024 --sort date

# Code search — find code across public repos (via Sourcegraph)
scripts/search github "func NewRydbergHamiltonian" --code --count 5
scripts/search github "class QuantumCircuit" --code

# Read README — quick overview without cloning
scripts/search github qiskit/qiskit --readme

# Clone repo — shallow clone to local dir for deep analysis
scripts/search github qiskit/qiskit --clone --output /tmp

Repo search (default): GitHub REST API. Sorted by stars (default) or --sort date. --from-year YYYY filters by creation date. Auth via GITHUB_TOKEN env var or gh auth token (optional, increases rate limit from 10 to 30 req/min).
Code search (--code): Sourcegraph streaming API. Supports regex. Searches across most popular public repos. No auth needed.
README (--readme): Fetches raw README markdown via GitHub API. Pass owner/repo as the query.
Clone (--clone): git clone --depth 1 (latest snapshot only, no history). Default output: /tmp. Pass owner/repo as the query.

URL Fetch

bash

scripts/search fetch "https://example.com/page"

Strips HTML to plain text. For Cloudflare-protected sites, use scripts/browse instead.

Show full SKILL.md (293 more words)Show less

scripts/browse

Browser automation via browser-use. Always launches in headed mode with the user's real Chrome profile (cookies, fingerprint), so it bypasses Cloudflare and Google Scholar anti-bot.

The daemon persists across commands (~50ms latency after first open). Closed automatically when Luxas exits.

bash

scripts/browse open "https://journals.aps.org/prl/"
scripts/browse state                           # clickable elements with indices
scripts/browse input 5 "Rydberg atom"          # fill search box by index
scripts/browse click 7                         # click element by index
scripts/browse eval "document.body.innerText"  # extract full page text
scripts/browse close                           # shutdown (optional, auto-closed on exit)

Workflow: extract text from a protected page

bash

scripts/browse open "https://journals.aps.org/prl/abstract/10.1103/PhysRevLett.117.073003"
# If cookie consent appears, click accept:
scripts/browse state                           # find the accept button index
scripts/browse click <accept_index>
# Extract content:
scripts/browse eval "document.querySelector('.abstract')?.innerText"
scripts/browse eval "document.querySelector('h3.title')?.innerText"

Figure Extraction

bash

scripts/extract-figures data/papers/2104.10350.pdf
scripts/extract-figures data/papers/2301.07041              # arXiv source dir
scripts/extract-figures data/papers/paper.pdf --output report/figures/extracted

arXiv source: Parses .tex for \includegraphics, copies original figure files (PDF/PNG/EPS). Best quality.
PDF: Renders pages at 200 DPI, detects "Figure N" captions, crops figure regions. Works on both raster and vector figures.
Outputs a manifest.json with figure metadata (filename, caption, page number).
Use this to extract key figures from papers for inclusion in your report.

Decision Guide

Use scripts/search first — it covers most academic needs without a browser:

Need	Command
Find papers by keyword	`search papers <query>`
Chase citation chains	`search citations <id>`
Download paper (arXiv)	`search download --arxiv <id>`
Download paper (DOI/Sci-Hub)	`search download --doi <doi>`
Read LaTeX source	`search source <arxivId>`
Get BibTeX citation	`search bib <doi>`
General web search	`search web <query>`
Fetch unprotected URL	`search fetch <url>`
Extract figures from paper	`extract-figures <paper-path>`
Find GitHub repos	`search github <query>`
Search code across repos	`search github <query> --code`
Read repo README	`search github owner/repo --readme`
Clone repo for analysis	`search github owner/repo --clone`

Use scripts/browse only when search cannot do the job — it launches a real browser which is heavier:

Need	Command
Government reports, policy docs	`browse open <url>` → `eval "document.body.innerText"`
Non-academic web content (news, industry)	`browse open <url>` → `state` → interact
Cloudflare-blocked page (last resort)	`browse open <url>` → extract content
Interactive site requiring login/forms	`browse open <url>` → `state` → `input`/`click`

Rule of thumb: search for papers, academic data, and GitHub repos/code; browse for everything else (government sites, general web, protected pages). Don't use browse for tasks that search can handle.

Files

SKILL.md and 7 other files (scripts) in skills/search of Muuuun/luxas.

SKILL.md
scripts/browse
scripts/extract-figures
scripts/merge-notes
scripts/sb-browser
scripts/scihub-discover
scripts/search
scripts/setup.sh

Open the folder on GitHubat commit 9f77cef

Compare with similar skills

Search 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.

Search compared with similar skills
Skill	Stars	Used in	Tokens	Auto-check	Licence	Repo updated
Search this skillMuuuun/luxas	1.2k	—	~2.3k	Automated safety check: Warn	MIT	1 mo ago
Arxiv MCP Serverblazickjp/arxiv-mcp-server	3.2k	—	~353	Automated safety check: Pass	Apache-2.0	yesterday
Arxiv Paper Writerappautomaton/latex-arxiv-SKILL	458	—	~2.3k	Automated safety check: Pass	MIT	26 days ago
Arxiv Paper Writeryunshenwuchuxun/latex-paper-skills	266	—	~3.2k	Automated safety check: Pass	MIT	6 mo ago
Latex Paper ArtifactBingHanOfUESTC/open_agent_team	106	—	~635	Automated safety check: Pass	MIT	3 mo ago
Write Paperfrenzymath/Danus	476	—	~18k	Automated safety check: Pass	Apache-2.0	1 mo ago

Similar skills

Arxiv MCP Server
blazickjp/arxiv-mcp-server
A skill your agent uses when finding, comparing, reading, or monitoring arXiv papers, including requests for abstracts, citation graphs, original LaTeX, section-level technical details, or…
3.2k GitHub stars~353 tokensUpdated yesterday
Research & ScienceAuto-check passed
Arxiv Paper Writer
appautomaton/latex-arxiv-SKILL
Write LaTeX ML/AI review articles for arXiv using the IEEEtran template and verified BibTeX citations.
458 GitHub stars~2.3k tokensUpdated 26 days ago
Research & ScienceAuto-check passed
Arxiv Paper Writer
yunshenwuchuxun/latex-paper-skills
Writes ML/AI review and survey papers for arXiv using the IEEEtran LaTeX template with verified BibTeX citations.
266 GitHub stars~3.2k tokensUpdated 6 mo ago
Research & ScienceAuto-check passed
Latex Paper Artifact
BingHanOfUESTC/open_agent_team
A skill your agent uses when producing the final arXiv-style LaTeX paper, BibTeX, figures, tables, reproducibility statement, appendix, and artifact manifest.
106 GitHub stars~635 tokensUpdated 3 mo ago
Documents & OfficeAuto-check passed
Write Paper
frenzymath/Danus
Turn a project's verified fact graph into a publishable LaTeX paper in a configurable house style — a standalone amsart .tex with a real bibliography, compiled to PDF.
476 GitHub stars~18k tokensUpdated 1 mo ago
Documents & OfficeAuto-check passed
Arxiv Preflight
Mathews-Tom/armory
Pre-submission validation audit for arXiv papers across TeX source, PDF, figures, metadata, bibliography, file organization, and common-error scans.
328 GitHub stars~2k tokensUpdated 4 days ago
Research & ScienceAuto-check passed

Questions about Search

What does Search do?

Unified academic paper search, citation chains, paper download (arXiv LaTeX/PDF, Sci-Hub), figure extraction from papers, LaTeX source reading, BibTeX fetching, web search, and browser automation…. Search is an agent skill from Muuuun/luxas. Unified academic paper search, citation chains, paper download (arXiv LaTeX/PDF, Sci-Hub), figure extraction from papers, LaTeX source reading, BibTeX fetching, web search, and browser automation for Cloudflare-protected sites (PRL, Science, Nature, Google Scholar).

When should I use Search?

Search fits situations like: tasks that involve Academic paper search; tasks that involve Citation management; tasks that involve LaTeX.

How do I install Search in Claude Code?

Run `npx skills add Muuuun/luxas --skill search -a claude-code`. Or copy the skill folder (skills/search in Muuuun/luxas) into .claude/skills/search in your project. Claude Code loads it when a task matches its description.

How do I install Search in Codex?

Run `npx skills add Muuuun/luxas --skill search -a codex`. Or copy the skill folder (skills/search in Muuuun/luxas) into .agents/skills/search in your project. Codex loads it when a task matches its description.

Can I use Search 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 Muuuun/luxas --skill search -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/search, .gemini/skills/search, .github/skills/search and .opencode/skills/search in your project.

What does Search need to run?

Going by SKILL.md and its folder, Search needs a shell for the scripts in its folder, the command-line tools its instructions call (bash, gh and git) and credentials named BRAVE_API_KEY and GITHUB_TOKEN. Our summary lists: Node.js; A Bash shell; A credential in BRAVE_API_KEY; A credential in GITHUB_TOKEN. Its frontmatter pre-approves these tools: Bash(search:*), Bash(browse:*), Bash(browser-use:*), Bash(extract-figures:*). Compatibility (from SKILL.md): Requires Node.js 22+. Optional: browser-use CLI for browser automation. PyMuPDF and Pillow for figure extraction..

Does Search access the network?

SKILL.md names 1 domain. In commands or code: journals.aps.org; the agent is likely to contact it when it follows the instructions. This is read from the text; nothing was executed.

Is Search safe to install?

Our automated static check of SKILL.md flagged 1 warning(s): mentions a credentials file (ssh keys, cloud or package-manager tokens). Read the flagged lines before installing; the check is not a guarantee either way. The check reads SKILL.md only: the scripts in the folder are not scanned, so read them before running anything.

What licence does Search use?

Search is published under the MIT licence (the repository's licence). It allows redistribution, so the full SKILL.md is shown on this page.

How many tokens does Search use?

About 2.3k tokens (SKILL.md is roughly 9.4k characters). Agents keep only the skill's name and description in context until a task matches; then they load SKILL.md in full.

What are the alternatives to Search?

Skills that share tags, products or a category with Search: Arxiv MCP Server (blazickjp/arxiv-mcp-server, 3.2k stars), Arxiv Paper Writer (appautomaton/latex-arxiv-SKILL, 458 stars), Arxiv Paper Writer (yunshenwuchuxun/latex-paper-skills, 266 stars) and Latex Paper Artifact (BingHanOfUESTC/open_agent_team, 106 stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.

Who maintains Search?

Muuuun (a GitHub user) maintains it in Muuuun/luxas, which has 1,170 GitHub stars. The repository holds 9 skills in this directory. The repository was last updated on September 6, 2026.

Source: Muuuun/luxas on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.

Install Search

Facts

At a glance

What it does

When your agent uses it

Example prompts

Requirements

What it can do on your machine

Tool permissions

Runs code

Network

Credentials

Compatibility

Context cost

Safety

SKILL.md

Search Skill

Setup

scripts/search

Paper Search

Citation Chains

Paper Download

LaTeX Source

BibTeX

Web Search (Brave)

GitHub

URL Fetch

scripts/browse

Workflow: extract text from a protected page

Figure Extraction

Decision Guide

Files

Compare with similar skills

Similar skills

More from Muuuun/luxas

Related

Questions about Search