Documentation
Four agents, four endpoints. Here's how to use them.
Quick Start
Get API Key
Create an account and generate your API key from the dashboard.
Make Request
Send a POST request with your API key in the header.
Get Results
Receive structured JSON with the agent's response.
Authentication
All requests require an API key. Include it in one of these headers:
x-api-key: your_api_keyAuthorization: Bearer your_api_keyNote: Keep your API key secure. Never expose it in client-side code.
Search Agent
POSTSearches the web and returns structured results with sources.
/api/searchRequest Body
{
"messages": [
{
"role": "user",
"content": "Latest developments in AI"
}
]
}Response
{
"text": "Here are the latest AI developments...",
"sources": [
"https://example.com/ai-news",
"https://example.com/research"
]
}cURL Example
curl -X POST https://api.scira.ai/api/search \
-H "Content-Type: application/json" \
-H "x-api-key: YOUR_API_KEY" \
-d '{
"messages": [{
"role": "user",
"content": "Latest developments in AI"
}]
}'People Agent
POSTFinds and enriches people profiles across LinkedIn, GitHub, X, and more.
/api/peopleRequest Body
{
"query": "Andrej Karpathy AI researcher"
}Response
{
"profile": {
"name": "Andrej Karpathy",
"x": "https://x.com/karpathy",
"linkedin": "https://linkedin.com/in/andrej-karpathy",
"github": "https://github.com/karpathy",
"website": "https://karpathy.ai/",
"bio": "AI researcher, former Tesla AI director..."
}
}cURL Example
curl -X POST https://api.scira.ai/api/people \
-H "Content-Type: application/json" \
-H "x-api-key: YOUR_API_KEY" \
-d '{
"query": "Andrej Karpathy AI researcher"
}'Tip: Include context like company, role, or field for better results.
X Agent
POSTSearches X in real-time. Optionally target a specific user.
/api/xsearchRequest Body
{
"query": "AI Gateway",
"username": "vercel"
}username is optional. Omit to search all of X.
Response
{
"text": "## AI Gateway\n\n...markdown answer with inline citations...",
"sources": [
"https://x.com/vercel/status/123...",
"https://x.com/vercel/status/456..."
],
"results": [
{
"url": "https://x.com/vercel/status/123...",
"title": "Tweet title",
"text": "Tweet content from Exa...",
"author": "vercel",
"publishedDate": "2024-01-15T10:30:00.000Z",
"tweet": {
"id": "123...",
"text": "Full tweet text...",
"created_at": "Mon Jan 15 10:30:00 +0000 2024",
"user": {
"name": "Vercel",
"screen_name": "vercel",
"profile_image_url": "https://pbs.twimg.com/...",
"verified": true
},
"favorite_count": 1234,
"conversation_count": 56,
"photos": [],
"video": null
}
}
]
}Each result includes enriched tweet data from the X API when available. The tweet field may be null if the tweet could not be fetched.
cURL Example
curl -X POST https://api.scira.ai/api/xsearch \
-H "Content-Type: application/json" \
-H "x-api-key: YOUR_API_KEY" \
-d '{
"query": "AI Gateway",
"username": "vercel"
}'Reddit Agent
POSTSearches Reddit discussions and returns an agent summary with citations plus the underlying results.
/api/redditRequest Body
{
"query": "Best practices for deploying Next.js on Vercel?"
}Response
{
"text": "## Answer\n\n...markdown answer with inline citations...",
"sources": [
"https://www.reddit.com/r/...",
"https://www.reddit.com/r/..."
],
"results": [
{
"title": "Thread title",
"url": "https://www.reddit.com/r/...",
"excerpt": "Relevant excerpt..."
}
]
}cURL Example
curl -X POST https://api.scira.ai/api/reddit \
-H "Content-Type: application/json" \
-H "x-api-key: YOUR_API_KEY" \
-d '{
"query": "Best practices for deploying Next.js on Vercel?"
}'Errors
Standard HTTP status codes with JSON error messages.
API key missing or invalid.
Rate limit or quota exceeded.
Invalid request format or missing required fields.
Something went wrong on our end.
Error Response Format
{
"error": "API key is required."
}