Discovery

The discovery endpoint returns the complete Proxagora API catalog. It's the starting point for any agent that needs to find and call external data APIs.

Endpoint

GET https://proxagora.com/api/discover

No authentication required. No payment. Free catalog access.

Response

{
  "apis": [
    {
      "id": "ip-geo",
      "name": "IP Geo",
      "category": "Network",
      "cost_usdc": 0.001,
      "description": "Geographic location, ASN, and carrier data for any IPv4/IPv6 address",
      "endpoint": "/api/ip-geo"
    },
    {
      "id": "domain-reputation",
      "name": "Domain Reputation",
      "category": "Domain",
      "cost_usdc": 0.005,
      "description": "Risk score, blacklist status, and threat intelligence for any domain"
    },
    {
      "id": "email-validation",
      "name": "Email Validation",
      "category": "Contact",
      "cost_usdc": 0.002,
      "description": "Deliverability check, MX validation, and disposable domain detection"
    }
  ],
  "total": 32
}

API Object Fields

| Field | Type | Description | |-------|------|-------------| | id | string | Unique identifier — used as {id} in call endpoint | | name | string | Human-readable name | | category | string | Category: Domain, Network, Contact, Tech, Intelligence, Business, Media, Travel | | cost_usdc | number | Per-call cost in USDC | | description | string | What the API returns | | endpoint | string | Full path (same as /api/{id}) |

Filtering

By Category

curl "https://proxagora.com/api/discover?category=Domain"

Available categories: Domain, Network, Contact, Tech, Intelligence, Business, Media, Travel

By Max Cost

curl "https://proxagora.com/api/discover?max_cost=0.005"

Returns only APIs at or below the specified USDC cost.

Combined Filters

curl "https://proxagora.com/api/discover?category=Contact&max_cost=0.01"

Current API Catalog

| ID | Name | Category | Cost | |----|------|----------|------| | ip-geo | IP Geo | Network | $0.001 | | whois | WHOIS | Domain | $0.001 | | weather | Weather | Travel | $0.001 | | github-stats | GitHub Stats | Tech | $0.001 | | youtube-metadata | YouTube Metadata | Media | $0.002 | | email-validation | Email Validation | Contact | $0.002 | | news-search | News Search | Media | $0.005 | | domain-reputation | Domain Reputation | Domain | $0.005 | | tech-stack | Tech Stack | Tech | $0.005 | | hackernews-search | HackerNews Search | Media | $0.005 | | reddit-search | Reddit Search | Media | $0.005 | | wikipedia-summary | Wikipedia Summary | Intelligence | $0.005 | | webpage-content | Webpage Content | Intelligence | $0.005 | | phone-validation | Phone Validation | Contact | $0.002 | | vat-validation | VAT Validation | Business | $0.002 | | exchange-rates | Exchange Rates | Business | $0.001 | | serp-results | SERP Results | Intelligence | $0.01 | | email-finder | Email Finder | Contact | $0.01 | | linkedin-enrichment | LinkedIn Enrichment | Business | $0.02 | | url-screenshot | URL Screenshot | Tech | $0.01 | | youtube-transcript | YouTube Transcript | Media | $0.005 | | reddit-thread | Reddit Thread | Media | $0.005 |

Using Discovery in Agent Code

import requests

def discover_apis(category: str = None, max_cost: float = None) -> list[dict]:
    params = {}
    if category:
        params["category"] = category
    if max_cost is not None:
        params["max_cost"] = max_cost
    
    response = requests.get("https://proxagora.com/api/discover", params=params)
    return response.json()["apis"]

# Find all cheap Network APIs
network_apis = discover_apis(category="Network", max_cost=0.005)

# Let an LLM pick the right API for a task
import json
catalog = discover_apis()
catalog_str = json.dumps([{"id": a["id"], "description": a["description"]} for a in catalog])

# Include in LLM system prompt or tool definition

const catalog = await fetch('https://proxagora.com/api/discover')
  .then(r => r.json())
  .then(d => d.apis)

// Find API by name
const ipGeo = catalog.find((a: any) => a.id === 'ip-geo')
console.log(`Call /api/${ipGeo.id} — costs $${ipGeo.cost_usdc} USDC`)

Caching

The catalog changes infrequently (new APIs added, existing ones updated). For production agents:

• Cache the response for 5–60 minutes
• Refresh on startup
• Don't call discover on every API call — it's expensive in latency

import time

_catalog_cache = None
_cache_time = 0
CACHE_TTL = 300  # 5 minutes

def get_catalog():
    global _catalog_cache, _cache_time
    if _catalog_cache is None or time.time() - _cache_time > CACHE_TTL:
        _catalog_cache = discover_apis()
        _cache_time = time.time()
    return _catalog_cache