菜单
Comprehensive Exa.ai skill - semantic search, Research API, Answer endpoint, Websets, LinkedIn discovery, and company research
用 Codex 或 Claude 帮你安装 复制这段 Prompt，粘贴到 Codex、Claude 或其他助手里，让它检查 Skill 页面并帮你完成安装。
基于 SOC 职业分类
| name | exa |
| description | Comprehensive Exa.ai skill - semantic search, Research API, Answer endpoint, Websets, LinkedIn discovery, and company research |
Use this skill when working with Exa.ai - the embeddings-based semantic search engine. This skill covers all Exa products and APIs.
| Endpoint | Purpose | Cost |
|---|---|---|
/search | Core search (4 modes: fast/auto/neural/deep) | $0.005-0.015 |
/contents | Get page text/highlights/summary | $0.001/page |
/findSimilar | Find pages similar to URL | $0.005 |
/answer | LLM answers with citations | $0.005 |
/research/v1 | Async agentic research | $0.005/task + pages |
outputSchema| Operation | Cost |
|---|---|
| Search (1-25 results, fast/auto/neural) | $0.005 |
| Search (26-100 results) | $0.025 |
| Deep search (1-25 results) | $0.015 |
| Content text per page | $0.001 |
| Content highlights per page | $0.001 |
| Content summary per page | $0.001 |
| Answer | $0.005 |
| Research task | $0.005 + $0.005-0.01/page |
Typical company research: $0.01-0.02
Claude Code has built-in Exa MCP tools:
| Tool | Purpose |
|---|---|
mcp__exa__web_search_exa | Web search with configurable results |
mcp__exa__get_code_context_exa | Get code/documentation context |
# Find a person's LinkedIn
/exa "John Smith CEO TechCorp" linkedin
# Research a company
/exa "Anthropic AI safety company overview"
# Find competitors
/exa similar:https://anthropic.com
# Quick factual answer
/exa answer:"When was OpenAI founded?"
# Deep research (async)
/exa research:"Find 10 AI startups in healthcare Series A 2024"
For complete API documentation, parameters, and code examples:
references/exa-complete-reference.md
Covers:
Walk any independent professional — consultant, fractional executive, agency owner, solo founder, advisor — from a fuzzy "I help everyone" identity to an ultra-specific niche AND a pain-sorted list of real prospects with decision-maker contacts and tailored openers. This is the full niche-discovery → testable-ICP → prospecting pipeline (the same one built for fractional CTO Carl Maloney in this repo). Trigger this skill whenever the user says "who do I want to be when I grow up", "who should I be when I grow up", "help me find my niche", "I don't know who my customer is", "who exactly should I sell to", "help me figure out my positioning / ICP", "who needs what I do right now", "build me a target list for my consulting/fractional/agency practice", or any variant where someone selling their own expertise needs to (a) figure out their edge and (b) find who's in the situation that needs it. Use it even when the person doesn't explicitly say "niche" — if they're an independent professional unsure of their positio
Find the best data sources for any vertical. Describe your niche, ICP, or industry and this skill discovers the public + commercial data sources that reveal which companies are in pain (a trigger or problem), verifies the actual fields exist, and scores every source on 7 criteria — quality, scale, price, freshness, accessibility, granularity, and pain-signal strength — ranking cheap + high-quality first. Use when someone asks "what data exists for [vertical]", "where can I find a list of [company type]", "how do I find companies with [problem/trigger]", "best data source for [industry]", or wants to build a targeting / prospecting dataset.