Exa Web Search Free
Run Exa-powered web and code-context searches from your agent via mcporter when you need fresh sources instead of stale training data.
Overview
Exa Web Search Free is an agent skill most often used in Idea (also Validate, Build) that teaches mcporter call patterns for Exa web search and code-context retrieval.
Install
npx skills add https://github.com/sundial-org/awesome-openclaw-skills --skill exa-web-search-freeWhat is this skill?
- Web search via exa.web_search_exa with fast and deep type modes and configurable numResults
- Code-context retrieval via exa.get_code_context_exa with tokensNum budgets (e.g. 2000–5000)
- Ready mcporter call patterns for news, RAG/LLM topics, product comparisons, and framework docs
- Separate flows for general web research versus programming-specific context harvesting
- Example web searches use numResults values such as 5 and 8
- Code-context examples document tokensNum budgets including 2000, 3000, 4000, and 5000
Adoption & trust: 2.6k installs on skills.sh; 609 GitHub stars; 1/3 security scanners passed (skills.sh audits).
What problem does it solve?
You need timely web answers and snippet-level code documentation inside an agent session but lack a repeatable Exa invocation recipe.
Who is it for?
Solo builders running OpenClaw-style agent stacks who already use mcporter and want Exa search wired into research and doc-lookup loops.
Skip if: Teams that need a hosted search UI, strict offline-only workflows, or guaranteed zero network egress without reviewing agent permissions.
When should I use this skill?
You need Exa-backed web search or code-context examples invoked through mcporter during agent research or implementation.
What do I get? / Deliverables
You copy vetted mcporter commands that return ranked web results or token-bounded code context so the agent can cite fresher sources and implementation patterns in the next planning or coding step.
- Executable mcporter command templates for web and code-context searches
- Parameterized query patterns agents can reuse across research tasks
Recommended Skills
Journey fit
Spans multiple journey phases - primary shelf plus alternate fits below.
Canonical shelf is Idea → research because the skill’s primary value is discovering news, competitors, and topic depth before you commit to build. Examples center on web_search_exa and get_code_context_exa for open-ended discovery and documentation-style lookups, which maps directly to the research subphase.
Where it fits
Run deep web_search_exa queries on AI breakthroughs and quantum computing news before picking a niche.
Compare vector databases and Mac hardware specs with structured numResults to narrow MVP dependencies.
Pull get_code_context_exa snippets for Stripe checkout or Express error handling while wiring payments.
Fetch asyncio or API-design best-practice context to guide backend implementation choices.
How it compares
Use as an integration playbook for Exa via mcporter—not a self-contained local search index or generic browser automation skill.
Common Questions / FAQ
Who is exa-web-search-free for?
It is for solo and indie builders using agent tooling who want Exa web and code-context search callable from the terminal or agent via mcporter.
When should I use exa-web-search-free?
Use it during Idea research for markets and news, during Validate to compare products and approaches, and during Build when you need framework or API examples pulled into the agent context.
Is exa-web-search-free safe to install?
It implies network calls to Exa through mcporter; review the Security Audits panel on this Prism page and your agent network policy before enabling it in production repos.
SKILL.md
READMESKILL.md - Exa Web Search Free
# Exa Search Examples ## Web Search Examples ### Latest News & Current Events ```bash mcporter call 'exa.web_search_exa(query: "latest AI breakthroughs 2026", numResults: 5)' mcporter call 'exa.web_search_exa(query: "quantum computing news", type: "fast")' ``` ### Research Topics ```bash mcporter call 'exa.web_search_exa(query: "how does RAG work in LLMs", type: "deep", numResults: 8)' mcporter call 'exa.web_search_exa(query: "best practices for API design", numResults: 5)' ``` ### Product Information ```bash mcporter call 'exa.web_search_exa(query: "M4 Mac Mini specifications and reviews")' mcporter call 'exa.web_search_exa(query: "comparison of vector databases", type: "deep")' ``` ## Code Context Search Examples ### Programming Language Basics ```bash mcporter call 'exa.get_code_context_exa(query: "Python asyncio basics and examples", tokensNum: 3000)' mcporter call 'exa.get_code_context_exa(query: "Rust ownership and borrowing tutorial")' ``` ### Framework & Library Usage ```bash mcporter call 'exa.get_code_context_exa(query: "React useState and useEffect hooks examples", tokensNum: 2000)' mcporter call 'exa.get_code_context_exa(query: "Next.js 14 app router authentication middleware")' mcporter call 'exa.get_code_context_exa(query: "Express.js error handling best practices", tokensNum: 4000)' ``` ### Specific API & SDK Documentation ```bash mcporter call 'exa.get_code_context_exa(query: "Stripe checkout session implementation", tokensNum: 5000)' mcporter call 'exa.get_code_context_exa(query: "AWS S3 SDK upload examples Python")' mcporter call 'exa.get_code_context_exa(query: "Discord.js bot slash commands")' ``` ### Debugging & Solutions ```bash mcporter call 'exa.get_code_context_exa(query: "fixing CORS errors in Node.js Express")' mcporter call 'exa.get_code_context_exa(query: "pandas dataframe memory optimization techniques", tokensNum: 4000)' ``` ## Company Research Examples ### Startups & Tech Companies ```bash mcporter call 'exa.company_research_exa(companyName: "Anthropic", numResults: 3)' mcporter call 'exa.company_research_exa(companyName: "Perplexity AI")' mcporter call 'exa.company_research_exa(companyName: "Scale AI", numResults: 5)' ``` ### Public Companies ```bash mcporter call 'exa.company_research_exa(companyName: "Microsoft")' mcporter call 'exa.company_research_exa(companyName: "NVIDIA", numResults: 5)' ``` ### Research Queries ```bash # Find funding info mcporter call 'exa.company_research_exa(companyName: "OpenAI", numResults: 5)' # Recent news mcporter call 'exa.company_research_exa(companyName: "Tesla", numResults: 3)' ``` ## Parameter Guidance ### `type` parameter (web_search_exa) - `"auto"` - Balanced search (default) - `"fast"` - Quick results, less comprehensive - `"deep"` - Thorough research, slower but more complete ### `tokensNum` parameter (get_code_context_exa) - `1000-2000` - Focused queries, specific examples - `3000-5000` - Standard documentation lookup (default: 5000) - `5000-10000` - Comprehensive guides and tutorials - `10000-50000` - Deep dives, full API documentation ### `numResults` parameter - `3-5` - Quick lookup, specific answer - `5-8` - Standard research (default for web: 8, company: 5) - `10+` - Comprehensive research, multiple perspectives ## Advanced Tools Examples (Off by Default) ### Deep Search ```bash # Comprehensive research mcporter call 'exa-full.deep_search_exa(query: "AI safety alignment research comprehensive overview")' # Multi-perspective exploration mcporter call 'exa-full.deep_search_exa(query: "climate change solutions technology innovation")' ``` ### Advanced Web Search ```bash # Search with domain filters mcporter call 'exa-full.web_search_advanced_exa(query: "machine learning tutorials", includeDomains: ["github.com", "arxiv.org"])' # Search with date range mcporter call 'exa-full.web_search_advanced_exa(query: "AI developments", startPublishedDate: "2026-01-01")' ``` ### Crawling ```bash # Extract content from specific URL mcporter