perplexity-web-research
Leverage Perplexity CLI for web research, deep reasoning, and independent validation. Use when you need current/external knowledge, want a second opinion before decisions, or want to delegate complex analytical thinking to specialized reasoning models. Complements (not replaces) standard web tools. Can be invoked directly or spawned as a subagent.
Profile →npx skills add majiayu000/claude-skill-registry --skill perplexity-web-research --agent claude-code
Same command for any agent — swap --agent for codex, cursor, copilot.
Weekly change comes from our own snapshots, not the repository page — it measures attention, not adoption.
# Perplexity Web Research Use the `llm` CLI with Perplexity plugin for web research, deep reasoning, and independent validation. ## Quick Start ```bash llm -m sonar 'your question here' ``` ## Model Selection Choose the appropriate model based on query complexity: | Model | Use Case | Context | Web Search | |-------|----------|---------|------------| | `sonar` | Quick lookups, general questions | 128k | ✓ | | `sonar-pro` | Thorough answers, detailed research (flagship) | 200k | ✓ | | `sonar-deep-research` | Complex multi-faceted research | 128k | ✓ | | `sonar-reasoning` | Step-by-step reasoning questions | 128k | ✓ | | `sonar-reasoning-pro` | Advanced reasoning with more detail | 128k | ✓ | | `r1-1776` | Specialized model without web search | 128k | ✗ | **Default choice**: `sonar` for most queries. **Use `sonar-pro`** when: - Query requires comprehensive coverage - Need detailed explanations - Research involves multiple aspects - Largest context window needed (200k) **Use `sonar-deep-research`** when: - Topic is complex and multi-faceted - Need synthesis across many sources - Research question has no simple answer **Use `sonar-reasoning`** when: - Question requires logical deductio
- Quick Start
- Model Selection
- When to Use
- 1. Current or External Knowledge
- 2. Second Opinions
- 3. Deep Reasoning Delegation
- When NOT to Use
- Usage Examples
- Current Knowledge: Documentation Lookup
- Current Knowledge: Error Research
- Current Knowledge: Library Version
- Current Knowledge: Complex Research
- Second Opinion: Architecture Decision
- Second Opinion: Technology Choice
llm -m sonar 'your question here' llm -m sonar 'Next.js App Router documentation 2026' llm -m sonar 'ECONNREFUSED error when connecting to PostgreSQL' llm -m sonar 'latest stable version of React 2026' llm -m sonar-deep-research 'comparison of WebSocket vs Server-Sent Events for real-time updates in 2026' llm -m sonar-pro 'trade-offs between microservices and monolith for a team of 5 engineers' llm -m sonar 'should I use PostgreSQL or MongoDB for a social media analytics platform' llm -m sonar-reasoning 'how to design an efficient caching strategy for a rate-limited API with bursty traffic patterns' llm -m sonar-reasoning-pro 'design a scalable job queue system that handles both immediate and scheduled tasks with priority levels'
What does the perplexity-web-research skill do?
Leverage Perplexity CLI for web research, deep reasoning, and independent validation. Use when you need current/external knowledge, want a second opinion before decisions, or want to delegate complex analytical thinking to specialized reasoning models. Complements (not replaces) standard web tools. Can be invoked directly or spawned as a subagent.
How do I install it?
Run `npx skills add majiayu000/claude-skill-registry --skill perplexity-web-research --agent claude-code` — it drops the skill into your project so the agent can pick it up. Swap the --agent value for codex, cursor or copilot if you use one of those.
Where does this skill come from?
From majiayu000/claude-skill-registry, a repository with 534 stars. We read it straight from the repository tree rather than a submitted listing, so what you see here is what is actually published.
Is a popular skill a good skill?
Not necessarily. Stars measure attention, not adoption — a repository can trend for a week and be abandoned. That is why we show the weekly change from our own snapshots next to the total, instead of a single flattering number.