agentleFS
Sign inStart free
tesla3tesla3/my-pi-skills/brave-search/SKILL.md

brave-search

Skill1 starsChanged 9 months agoMITView on GitHub

Web search and content extraction via Brave Search API. Use for searching documentation, facts, or any web content. Lightweight, no browser required.

  • Reads credentials
  • Installs packages

6 other projects use a version of this, 4 of them edited.

Adopted by

Most recently taken up by

  • Other versions6 projects use a version of this.

    Word for word · 0

    None found.

    Edited · 4

    1. badlogic/pi-skills★ 2.5k21% same
    2. ferologics/pi-skills★ 1421% same
    3. cadecode/agent-skills★ 116% same
    4. sarfraznawaz2005/agent-skills-collection★ 121% same

    Wrote their own version · 2

    1. aiagenta2z/onekey-gateway★ 52same name
    2. bushuhui/skills★ 2same name

    How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.

  • In this repository3 other files it gives its agents.

The file

We split this skill into the rules it gives an agent

1 rule. 0 are common, 0 are distinctive, and 1 is very specific to this project.

  • 0

    Common

    Most of the 5 edited versions we read keep these.

  • 0

    Distinctive

    This author's own choices. Worth a look.

  • 1

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Very specific:

    Use --preset for common filtering patterns.

    line 32

“Common” is measured: kept by at least 6 in 10 of the 5 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Brave Search

Which Tool

NeedTool
Search + read content (default)llm-context.js
Just URLs and snippetssearch.js
Fetch a specific known URLcontent.js

LLM Context (Default)

Single API call — searches the web AND returns pre-extracted content optimized for LLMs.

{baseDir}/llm-context.js "query"                                    # Standard
{baseDir}/llm-context.js "query" --preset code                      # Filter SEO spam for programming
{baseDir}/llm-context.js "query" --preset research                  # Boost academic/quality sources
{baseDir}/llm-context.js "query" --preset docs                      # Only official documentation
{baseDir}/llm-context.js "query" --tokens 16384 --urls 30           # Deep research
{baseDir}/llm-context.js "query" --freshness pw                     # Last week only
{baseDir}/llm-context.js "query" --goggles '$discard,site=x.com'    # Custom inline rules

Presets

Use --preset for common filtering patterns. These are built-in inline goggles rules — tested and working.

PresetEffect
codeDiscard 8 SEO farms (w3schools, geeksforgeeks, tutorialspoint, etc.), boost SO + /docs/ paths
researchBoost arxiv, ACM, HN, lobste.rs; downrank medium; discard SEO farms
docsAllow-list only: official docs (Python, MDN, Rust, Go, TS, Node), realpython, SO, GitHub. Everything else discarded.

Options

  • --tokens <num> — Max tokens (default: 8192, min: 1024, max: 32768)
  • --urls <num> — Max URLs (default: 20, max: 50)
  • --count <num> — Search results to consider (default: 20, max: 50)
  • --threshold <mode> — Relevance: strict, balanced (default), lenient, disabled
  • --country <code> — Two-letter country code (default: US)
  • --freshness <period> — pd (day), pw (week), pm (month), py (year)
  • --preset <name> — Built-in goggles: code, research, docs
  • --goggles <val> — Hosted goggle URL or custom inline rules (use \n between rules). Cannot combine with --preset.

Custom Goggles

For per-query customization beyond presets. See goggles-reference.md for full syntax, hosted goggles catalog, and domain lists.

# Hosted goggle (189 copycat scrapers removed)
{baseDir}/llm-context.js "query" --goggles 'https://raw.githubusercontent.com/brave/goggles-quickstart/main/goggles/copycats_removal.goggle'

# Hosted goggle (allow-list: only HN-popular domains, everything else discarded)
{baseDir}/llm-context.js "query" --goggles 'https://raw.githubusercontent.com/brave/goggles-quickstart/main/goggles/hacker_news.goggle'

# Inline: discard specific domains
{baseDir}/llm-context.js "query" --goggles '$discard,site=pinterest.com\n$discard,site=quora.com'

# Inline: allow-list pattern
{baseDir}/llm-context.js "query" --goggles '$discard\n$boost=5,site=arxiv.org\n$boost=1,site=github.com'

Key rules: $boost=N,site=domain (N=1-10), $downrank=N,site=domain, $discard,site=domain, $discard (discard all unmatched). Precedence: discard > boost > downrank.

Search

Returns URLs, snippets, and extra snippets.

{baseDir}/search.js "query"                          # 5 results
{baseDir}/search.js "query" -n 10                    # More results (max 20)
{baseDir}/search.js "query" --content                # Include page content as markdown
{baseDir}/search.js "query" --freshness pw           # Last week
{baseDir}/search.js "query" --goggles 'https://raw.githubusercontent.com/brave/goggles-quickstart/main/goggles/copycats_removal.goggle'

Options: -n <num>, --content, --country <code>, --freshness <period>, --goggles <url|rules>

Both hosted goggles (pass URL) and inline rules work with --goggles. See goggles-reference.md for hosted goggles catalog.

Extract Page Content

{baseDir}/content.js https://example.com/article

Setup

Requires BRAVE_API_KEY env var. Get one at https://api-dashboard.search.brave.com — $5/1k requests, $5 free monthly credits.

export BRAVE_API_KEY="your-key"    # Add to ~/.zprofile
cd {baseDir} && npm install         # Run once

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

Check your setup, free