perplexity

Skill: perplexity — Web Research via Perplexity API

When to Use

Use this skill when you need current, cited web information:

• Library version history and changelogs
• CVE / security vulnerability lookups
• API documentation for external services
• Current best practices for a technology
• Competitive or market research
• Anything that requires an up-to-date source

Prerequisites

Set your Perplexity API key (get one at https://www.perplexity.ai/settings/api):

export PERPLEXITY_API_KEY="pplx-..."   # add to .env or shell profile

How to Query

Quick query (one-shot Python call)

import os, json, urllib.request

def perplexity_search(query: str, model: str = "sonar") -> dict:
    key = os.environ["PERPLEXITY_API_KEY"]
    payload = {
        "model": model,           # sonar (fast) | sonar-pro (deep, cited)
        "messages": [
            {"role": "system", "content": "Be precise and cite your sources."},
            {"role": "user", "content": query},
        ],
        "max_tokens": 1024,
        "return_citations": True,
    }
    req = urllib.request.Request(
        "https://api.perplexity.ai/chat/completions",
        data=json.dumps(payload).encode(),
        headers={
            "Authorization": f"Bearer {key}",
            "Content-Type": "application/json",
        },
    )
    with urllib.request.urlopen(req, timeout=30) as resp:
        return json.loads(resp.read())

result = perplexity_search("latest FastAPI security best practices 2025")
print(result["choices"][0]["message"]["content"])
# Citations are in result["citations"]

Run inline

python3 -c "
import os, json, urllib.request
key = os.environ.get('PERPLEXITY_API_KEY', '')
if not key: print('Set PERPLEXITY_API_KEY'); exit(1)
q = 'YOUR QUERY HERE'
payload = json.dumps({'model':'sonar','messages':[{'role':'user','content':q}],'max_tokens':512}).encode()
req = urllib.request.Request('https://api.perplexity.ai/chat/completions', data=payload,
    headers={'Authorization':f'Bearer {key}','Content-Type':'application/json'})
with urllib.request.urlopen(req, timeout=30) as r: d = json.loads(r.read())
print(d['choices'][0]['message']['content'])
"