Paid research assistant with real-time web search. Use when user asks "what's the latest", "current pricing for", "recent news about", "search the web for", "fact check this", "what's new in", or needs up-to-date information beyond training data.
Scanned 9/2/2026
Install to Claude Code
npx -y skills add majiayu000/claude-skill-registry --skill perplexity-grahama1970-agent-skills --agent claude-codeInstalls into .claude/skills of the current project.
Are you the author of Perplexity Grahama1970 Agent Skills?
Add the live security badge to your README — it updates automatically with every re-scan.
[](https://www.skillsdirectory.com/skills/majiayu000-perplexity-grahama1970-agent-skills-claude-skill-registry)More formats (shields.io, HTML) on the badges page.
---
name: perplexity
description: >
Paid research assistant with real-time web search. Use when user asks "what's the latest",
"current pricing for", "recent news about", "search the web for", "fact check this",
"what's new in", or needs up-to-date information beyond training data.
allowed-tools: Bash, Read
triggers:
- what's the latest
- current pricing
- recent news
- search the web
- fact check
- what's new in
- look up online
- research this topic
- paid search
metadata:
short-description: AI research with web search
provides:
- perplexity
composes:
- dogpile
- memory
- task-monitor
---
# Perplexity Skill
AI-powered research assistant with real-time web search capabilities.
## Prerequisites
- `PERPLEXITY_API_KEY` in `.env` (get from https://www.perplexity.ai/settings/api)
## When to Use
- Current information (news, recent releases, pricing)
- Technical questions requiring up-to-date docs
- Fact-checking and verification
- Research questions needing multiple sources
- Comparisons (libraries, tools, approaches)
## Quick Start
```bash
# Simple query (quick answer)
python .pi/skills/perplexity/perplexity.py ask "What's new in Python 3.12?"
# With citations (JSON output)
python .pi/skills/perplexity/perplexity.py research "ArangoDB vs Neo4j for knowledge graphs"
# With specific model
python .pi/skills/perplexity/perplexity.py ask --model large "Best practices for Lean4 proofs"
```
## Commands at a Glance
| Command | What it does | Notes |
|---------|--------------|-------|
| `ask` | Fast answer with short citations preview | Matches the Typer command implemented in `perplexity.py ask` |
| `research` | Full answer + citations JSON | Equivalent to calling `_research()` and prints JSON unless `--no-json` |
| `models` | Lists available model aliases → API IDs | Useful when wiring the skill into other agents |
Every CLI example **must** include the subcommand (`ask`, `research`, or `models`) because the Typer app requires it; omitting the subcommand will raise a usage error. The shipped SKILL docs now mirror the exact runner arguments, so you can copy/paste them without edits.
## CLI Usage
Two commands: `ask` (quick answer) and `research` (with citations):
```bash
# Quick answer
python .pi/skills/perplexity/perplexity.py ask [OPTIONS] "your question"
# Research with citations (JSON output)
python .pi/skills/perplexity/perplexity.py research [OPTIONS] "your question"
Options:
--model, -m small|large|huge (default: small, fast)
--system, -s Custom system prompt
--json/--no-json JSON output (research default: --json)
```
## Python API
```python
# Import the internal function
from perplexity import _research
# Research with citations
result = _research("What is BM25 scoring?", model="small")
print(result["answer"])
print(result["citations"])
```
Note: The CLI commands `ask` and `research` are the primary interface. For Python usage, use `_research()` directly.
## Shared Helpers
- `.pi/skills/dotenv_helper.py` is auto-imported so `.env` keys (e.g., `PERPLEXITY_API_KEY`) load without manual sourcing.
- `.pi/skills/json_utils.py` is available if you need to repair downstream JSON before writing it to files; the `research` command already emits valid JSON, so most callers can stream it directly.
## Models
| Model | API Name | Speed | Use Case |
|-------|----------|-------|----------|
| `small` | sonar | Fast | Quick lookups, simple questions |
| `large` | sonar-pro | Medium | Technical research, comparisons |
| `huge` | sonar-reasoning | Slow | Complex analysis, reasoning |
## Examples
### Technical Research
```bash
python .pi/skills/perplexity/perplexity.py ask "How to implement hybrid search with BM25 and vector similarity?"
```
### Current Information
```bash
python .pi/skills/perplexity/perplexity.py ask "Latest Anthropic Claude API changes 2024"
```
### Comparison
```bash
python .pi/skills/perplexity/perplexity.py ask --model large "sentence-transformers vs OpenAI embeddings for code search"
```
## vs Context7
| Perplexity | Context7 |
|------------|----------|
| General research | Library-specific docs |
| Real-time web search | Curated documentation |
| Any topic | Code libraries only |
| Synthesized answers | Raw doc snippets |
Use **Context7** for: specific library API syntax, exact function signatures
Use **Perplexity** for: research, comparisons, current information, general questions
Is this your skill, or is something wrong with this listing? Request removal or report an issue. Author removals are honored within 72 hours.
No comments yet. Be the first to comment!