-
Notifications
You must be signed in to change notification settings - Fork 1
Getting Started
SearchCans Agent Skills help AI agents turn live web data into structured, traceable work:
Search current SERPs → read important pages → produce evidence-led output.
Use these Skills for deep research, SEO/GEO planning, competitor analysis, RAG preparation, and webpage extraction audits.
Install all official SearchCans Skills:
npx skills add SearchCans/searchcans-skillsOr install one workflow:
npx skills add https://github.com/SearchCans/searchcans-skills --skill searchcans-deep-researchAvailable workflows:
searchcans-deep-researchsearchcans-serp-content-gapsearchcans-reader-seo-audit
Create or sign in to your SearchCans account:
Create an API key, then set it only in your execution environment.
macOS or Linux:
export SEARCHCANS_API_KEY="your-api-key"PowerShell:
$env:SEARCHCANS_API_KEY = "your-api-key"Never put an API key into a prompt, source file, JSON report, screenshot, or Git commit.
Visit Dashboard → Free Redemption Codes.
SearchCans releases a new redemption code every week. Each code adds 1,000 API credits, providing enough capacity for practical testing, research, and daily workflows.
Use when the answer needs current, verifiable web evidence.
Example prompt:
Use $searchcans-deep-research to compare the current US market positioning of leading SERP APIs.
The workflow requires a 3–5 question research plan, searches a selected market, reads selected source pages, and permits consequential claims only when supported by successfully extracted URLs.
Use when you need localized search-intent and competitor evidence for SEO or GEO planning.
Example prompt:
Use $searchcans-serp-content-gap to analyze the US Google SERP for “AI search API” and identify content opportunities.
The workflow returns organic competitors, People Also Ask questions, related searches, and other available SERP signals.
Use when you need to inspect a webpage, PDF, or Office document as an LLM-ready source.
Example prompt:
Use $searchcans-reader-seo-audit to inspect this page for Markdown extraction quality, canonical URL, H1s, metadata, and JSON-LD.
- Treat SERP snippets as discovery signals, not factual proof.
- Use Reader-extracted pages to support important claims.
- Start with the smallest research and source budget that answers the question.
- Start Reader extraction with standard mode before using a higher-cost proxy tier.
- Respect website terms, privacy, copyright, and applicable law.