skills.rest
MoizIbnYousaf
MoizIbnYousafCOMMUNITY
Explore all skills by this author →
Website𝕏@MoizIbnYousafGitHub

build-with-exa

Integrate Exa search, answer, agent, monitors, and websets APIs into applications.

31|5|Updated Mar 12, 2026
Search Tags:#semantic search#web search#sdk#api integration#research automation#exa#agent api
One-click install
npx skills add https://github.com/MoizIbnYousaf/marketing-cli --skill build-with-exa-moizibnyousaf
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: build-with-exa
Source: https://github.com/MoizIbnYousaf/marketing-cli/tree/main/skills/build-with-exa
Command: npx skills add https://github.com/MoizIbnYousaf/marketing-cli --skill build-with-exa-moizibnyousaf

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires exa-py, exa-js, and includes references (resource) components.

What problem does it solve? Developers integrating Exa's API platform often pick the wrong endpoint, misplace request parameters, or use deprecated fields, causing failed calls and wasted debugging time. This Skill provides endpoint selection guidance, correct request shapes, and SDK naming rules for every Exa surface. ## Core Features & Use Cases - Endpoint Decision Workflow: Routes tasks to the right Exa surface, defaulting to POST /search with type auto and highlights, and escalating to the Agent API only for multi-step research. - Request Shape References: On-demand reference files cover search, contents, answer, context, agent, monitors, websets, OpenAI-compatible endpoints, and SDK casing differences between exa-py and exa-js. - Anti-Pattern Corrections: Documents common mistakes such as nesting contents fields incorrectly, using deprecated parameters like useAutoprompt and numSentences, or inventing unsupported category values. - Use Case: When building an agent that needs fresh web data, ask for an Exa integration and receive a correct /search call with highlights, proper EXA_API_KEY authentication, and guidance on when to use /answer, /context, or /agent instead. ## Quick Start Ask the agent to write an Exa search integration in Python or TypeScript that retrieves recent articles on a topic with highlights enabled.

Frequently Asked Questions about build-with-exa

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I integrate Exa search into my application?▼

Start with POST /search using type auto and contents.highlights set to true, authenticating with the x-api-key header. Use the exa-py or exa-js SDKs for simpler integration, and only escalate to specialized endpoints when the task shape requires it.

Which Exa endpoint should I use for my use case?▼

Use /search for general retrieval, /contents when you already know the URLs, /answer for grounded answers with citations, /context for code retrieval, and /agent for async multi-step research. Default to /search for most new integrations.

What is the difference between exa-py and exa-js naming conventions?▼

The Python SDK uses snake_case parameters like num_results, output_schema, and max_characters, while the TypeScript SDK uses camelCase like numResults and outputSchema. Raw HTTP requests also use camelCase field names.

Why does my Exa search request fail with parameter errors?▼

Common causes include placing text, highlights, or summary at the top level instead of inside contents on /search, using deprecated fields like useAutoprompt or numSentences, or inventing unsupported category values like github or pdf.

When should I use the Exa Agent API instead of search?▼

Use the Agent API for asynchronous multi-step research, list-building, and enrichment tasks that need structured output over many sources. It is higher-latency and polled asynchronously, so prefer /search for simple low-latency retrieval.

Does Exa support OpenAI SDK compatibility?▼

Yes, Exa exposes /chat/completions and /responses endpoints that accept OpenAI SDK clients by pointing base_url at https://api.exa.ai. Use extra_body for Exa-specific fields, though native endpoints are preferred for new integrations.

Related Skills

Explore More Software Engineering Skills→
sputnikussputnikus

exa-search

Search the web and retrieve information using the Exa API.

Community
Intermediate
JhabbigJhabbig

rag-implementation

Build retrieval-augmented generation systems that ground LLM answers in external knowledge.

Community
Advanced
alin-blipalin-blip

AI Sales Team — Main Orchestrator

Orchestrate AI sales agents for prospect analysis, lead qualification, and outreach generation.

Community
Advanced
L-LesterYuL-LesterYu
51

tavily-search-zh

Perform web searches via the Tavily API and retrieve titles, URLs, and content summaries.

Community
Basic