convez.ai
MCP Server
Crawlora-org

crawlora-skills

Crawlora-org/crawlora-skills

Ver no GitHub
InicialPluginScore · de 0 a 100

Crawlora Agent Skills — structured public web data (search, e-commerce, social, finance, app stores, media, reviews) via the Crawlora REST API. Installable in any agent with the skills CLI or as a Claude Code plugin marketplace.

Instalação

$npx convez-ai Crawlora-org/crawlora-skills

1 comando no terminal: detecta o formato do repositório (marketplace ou plugin avulso) e instala via Claude Code. Requer Node 18+ e o Claude Code instalado.

Métricas

Estrelas
1
Forks
0
Acompanhando
0
Saúde
50%
Último commit
há 2 dias
Versão
—
Issues abertas
0
Contribuidores
1

README

A tradução para o português deste README ainda está na fila.

Crawlora Agent Skills

Website Docs Skills License: MIT

Crawlora Agent Skills are ready-to-install Agent Skills that teach any coding agent (Claude Code, Codex, Cursor, Copilot, …) how to fetch structured public web data with the Crawlora REST API — search engines, marketplaces, social, finance, maps, app stores, media, and reviews — getting clean JSON instead of HTML to parse.

Each skill is standalone REST: a tiny curl helper hitting https://api.crawlora.net/api/v1 with your CRAWLORA_API_KEY. No MCP setup required — install one skill and go. (If you'd rather use agent-native MCP tools, see crawlora-mcp.)

Get a free Crawlora API key — 2,000 credits/mo, no card — at crawlora.net. Billing is pay-on-success (only 2xx responses are charged).

The skills

SkillWhat it doesPlatforms
crawloraUmbrella catalog skill — fetch structured data from any of the 445 Crawlora endpoints; teaches auth, credits, and the full catalog.all 38
product-price-researchFind products, compare prices/sellers, pull marketplace reviews.Amazon, eBay, Shopify, Shop.app
youtube-researchTranscripts, comments, video/channel metadata, search — no yt-dlp.YouTube
app-review-miningApp details, reviews, ratings, rankings, similar apps (ASO).App Store, Google Play
serp-keyword-researchSERP snapshots, autocomplete keyword ideas, Google Trends.Google, Bing, Brave, Google Trends

Install

With the skills CLI (any agent)

Install one skill (or several, or all) straight from this repo:

# one skill
npx skills add github.com/Crawlora-org/crawlora-skills --skill youtube-research

# several
npx skills add github.com/Crawlora-org/crawlora-skills \
  --skill product-price-research --skill serp-keyword-research

# everything
npx skills add github.com/Crawlora-org/crawlora-skills --all

Then set your key once:

export CRAWLORA_API_KEY=sk_your_key_here

As a Claude Code plugin marketplace

This repo also ships a .claude-plugin/marketplace.json, so you can add it as a marketplace and install the bundle:

/plugin marketplace add Crawlora-org/crawlora-skills
/plugin install crawlora-skills@crawlora-skills

Usage

Every skill bundles a scripts/crawlora.sh helper and a reference/ file listing the exact endpoints, methods, and params it uses. The agent reads SKILL.md, picks the right endpoint, and calls it:

# GET (key=value params)
scripts/crawlora.sh /amazon/search k="standing desk" | jq '.'
scripts/crawlora.sh /youtube/transcript/dQw4w9WgXcQ | jq '.'

# POST (JSON body)
scripts/crawlora.sh -X POST /google/search '{"searchOption":{"q":"web scraping api"}}' | jq '.'

What's inside

crawlora-skills/
├── .claude-plugin/marketplace.json   # also a Claude Code plugin marketplace
├── lib/crawlora.sh                   # shared REST helper template (GET + POST)
├── scripts/
│   ├── tools.json                    # endpoint catalog (source of truth, from crawlora-mcp)
│   ├── generate.mjs                  # emits each skill's reference/ + bundled helper
│   └── validate.mjs                  # frontmatter + body-length lint
└── skills/<name>/{SKILL.md, scripts/crawlora.sh, reference/…}

The endpoint catalog comes from Crawlora's published API catalog (vendored as scripts/tools.json). Regenerate the per-skill references after a catalog change:

node scripts/generate.mjs        # write
node scripts/generate.mjs --check  # CI parity check
node scripts/validate.mjs        # lint all SKILL.md

Links

License

MIT. The Crawlora API itself is a hosted SaaS governed by the Crawlora terms; this repository contains the skill definitions, helper, and manifests.

Ver o original no GitHub

Bom para

Relacionados