curl "https://pt-edge.onrender.com/api/v1/quality?domain=agents&subcategory=autonomous-research-labs&limit=1"

Sample response

{
  "data": [{
    "full_name": "openai/openai-agents-python",
    "name": "openai-agents-python",
    "description": "A lightweight, powerful framework for multi-agent workflows",
    "stars": 19951, "forks": 3261,
    "language": "Python", "license": "MIT",
    "subcategory": "agent-framework-patterns",
    "pypi_package": "openai-agents",
    "downloads_monthly": 21224096,
    "quality_score": 98, "quality_tier": "verified",
    "maintenance_score": 25, "adoption_score": 25,
    "maturity_score": 25, "community_score": 23
  }],
  "meta": {
    "timestamp": "2026-04-11T12:00:00+00:00",
    "count": 1,
    "query": {"domain": "agents", "limit": 1}
  }
}

Quick Start

Pick your language. No signup needed for the first 100 requests/day.

Python

import requests

API = "https://pt-edge.onrender.com/api/v1"

# Quality scores — no auth needed
resp = requests.get(f"{API}/quality", params={"domain": "agents", "quality_tier": "verified", "limit": 5})
for p in resp.json()["data"]:
    print(f"{p['name']:30s}  score={p['quality_score']}  stars={p['stars']:,.0f}")

JavaScript

const API = "https://pt-edge.onrender.com/api/v1";

const resp = await fetch(`${API}/quality?domain=agents&quality_tier=verified&limit=5`);
const { data } = await resp.json();
data.forEach(p => console.log(`${p.name} — score=${p.quality_score}, stars=${p.stars.toLocaleString()}`));

curl

curl -s "https://pt-edge.onrender.com/api/v1/quality?domain=agents&quality_tier=verified&limit=5" | python3 -m json.tool

With an API key (1,000 requests/day):

Python

# With API key (1,000 requests/day)
headers = {"Authorization": "Bearer pte_YOUR_KEY"}
resp = requests.get(f"{API}/trending", headers=headers, params={"window": "7d", "limit": 10})

What's In The Data

PT-Edge tracks 220,000+ open-source AI repos across 30 domains. Each repo is scored 0–100 based on four components: maintenance (commit activity, issue response), adoption (downloads, dependents), maturity (age, stability, documentation), and community (stars, forks, contributors). Data flows in daily from GitHub, PyPI, npm, HuggingFace, and Hacker News.

Rate Limits

All tiers are free. All tiers get the same data. Resets at midnight UTC.

Get a Key

curl

# 1,000 requests/day — instant, no email
curl -X POST https://pt-edge.onrender.com/api/v1/keys

# 10,000 requests/day — just add your email
curl -X POST https://pt-edge.onrender.com/api/v1/keys \
  -H "Content-Type: application/json" \
  -d '{"email": "you@company.com"}'

Python

import requests
resp = requests.post("https://pt-edge.onrender.com/api/v1/keys")
key = resp.json()["data"]["key"]  # pte_...
print(f"Your key: {key}")

Then pass your key as a Bearer token: Authorization: Bearer pte_...

Core Endpoints

curl "https://pt-edge.onrender.com/api/v1/quality?domain=ml-frameworks&quality_tier=verified&limit=5"

curl "https://pt-edge.onrender.com/api/v1/trending?window=7d&domain=agents&limit=5"

curl "https://pt-edge.onrender.com/api/v1/projects/langchain"

curl "https://pt-edge.onrender.com/api/v1/datasets/quality?domain=mcp&quality_tier=verified&limit=100"

More Endpoints

Recipes

Copy-paste Python examples for common workflows.

Find the best tools in a domain

import requests
API = "https://pt-edge.onrender.com/api/v1"

# Top verified RAG frameworks
resp = requests.get(f"{API}/quality", params={"domain": "rag", "quality_tier": "verified", "limit": 10})
for p in resp.json()["data"]:
    print(f"{p['full_name']:40s}  score={p['quality_score']}  downloads={p['downloads_monthly']:,.0f}")

Track what's trending

# This week's fastest-growing projects
resp = requests.get(f"{API}/trending", params={"window": "7d", "limit": 10})
for p in resp.json()["data"]:
    print(f"{p['name']:25s}  +{p['stars_7d_delta']:,.0f} stars  lifecycle={p['lifecycle_stage']}")

Response Format

Every response returns a JSON envelope:

{
  "data": { ... },
  "meta": {
    "timestamp": "2026-04-03T12:00:00+00:00",
    "count": 20,
    "query": { "domain": "agents", "limit": 20 }
  }
}

Every response includes rate limit headers:

X-RateLimit-Limit: 100
X-RateLimit-Remaining: 99
X-RateLimit-Reset: 2026-04-04T00:00:00+00:00

Errors return structured JSON:

// 429 Rate limit
{ "detail": { "error": { "code": "rate_limit_exceeded", "message": "..." } } }

// 422 Invalid parameter
{ "detail": { "error": { "code": "invalid_domain", "message": "..." } } }

OpenAPI spec · AI plugin manifest · Atom feed