AI-powered web search, research, and reasoning via Perplexity
Install
npx skills add https://github.com/parcadei/continuous-claude-v3 --skill perplexity-searchSKILL.md
Perplexity AI Search
Web search with AI-powered answers, deep research, and chain-of-thought reasoning.
When to Use
- Direct web search for ranked results (no AI synthesis)
- AI-synthesized research with citations
- Chain-of-thought reasoning for complex decisions
- Deep comprehensive research on topics
Models (2025)
| Model | Purpose |
|---|---|
sonar |
Lightweight search with grounding |
sonar-pro |
Advanced search for complex queries |
sonar-reasoning-pro |
Chain of thought reasoning |
sonar-deep-research |
Expert-level exhaustive research |
Usage
Quick question (AI answer)
uv run python scripts/mcp/perplexity_search.py \
--ask "What is the latest version of Python?"
Direct web search (ranked results, no AI)
uv run python scripts/mcp/perplexity_search.py \
--search "SQLite graph database patterns" \
--max-results 5 \
--recency week
AI-synthesized research
uv run python scripts/mcp/perplexity_search.py \
--research "compare FastAPI vs Django for microservices"
Chain-of-thought reasoning
uv run python scripts/mcp/perplexity_search.py \
--reason "should I use Neo4j or SQLite for small graph under 10k nodes?"
Deep comprehensive research
uv run python scripts/mcp/perplexity_search.py \
--deep "state of AI agent observability 2025"
Parameters
| Parameter | Description |
|---|---|
--ask |
Quick question with AI answer (sonar) |
--search |
Direct web search - ranked results without AI synthesis |
--research |
AI-synthesized research (sonar-pro) |
--reason |
Chain-of-thought reasoning (sonar-reasoning-pro) |
--deep |
Deep comprehensive research (sonar-deep-research) |
Search-specific options
| Parameter | Description |
|---|---|
--max-results N |
Number of results (1-20, default: 10) |
--recency |
Filter: day, week, month, year |
--domains |
Limit to specific domains |
Mode Selection Guide
| Need | Use | Why |
|---|---|---|
| Quick fact | --ask |
Fast, lightweight |
| Find sources | --search |
Raw results, no AI overhead |
| Synthesized answer | --research |
AI combines multiple sources |
| Complex decision | --reason |
Chain-of-thought analysis |
| Comprehensive report | --deep |
Exhaustive multi-source research |
Examples
# Find recent sources on a topic
uv run python scripts/mcp/perplexity_search.py \
--search "OpenTelemetry AI agent tracing" \
--recency month --max-results 5
# Get AI synthesis
uv run python scripts/mcp/perplexity_search.py \
--research "best practices for AI agent logging 2025"
# Make a decision
uv run python scripts/mcp/perplexity_search.py \
--reason "microservices vs monolith for startup MVP"
# Deep dive
uv run python scripts/mcp/perplexity_search.py \
--deep "comprehensive guide to building feedback loops for autonomous agents"
API Key Required
Requires PERPLEXITY_API_KEY in environment or ~/.claude/.env.
Related skills
researchmattpocock575KInvestigate a question against high-trust primary sources and capture the findings as a Markdown file in the repo. Use when the user wants a topic researched, docs or API facts gathered, or reading legwork delegated to a background agent.paper-context-resolverlllllllama451KRigor Paper Context helper for README-first deep learning repo reproduction. Use only when the README and repository files leave a narrow reproduction-critical gap and the task is to resolve a specific paper detail such as dataset split, preprocessing, evaluation protocol, checkpoint mapping, or runtime assumption from primary paper sources while recording conflicts. Do not use for general paper summary, repo scanning, environment setup, command execution, title-only paper lookup, or replacing Renv-and-assets-bootstraplllllllama450KRigor Setup skill for README-first deep learning repo reproduction. Use when the task is specifically to prepare a conservative conda-first environment, checkpoint and dataset path assumptions, cache location hints, and setup notes before any run on a README-documented repository. Do not use for repo scanning, full orchestration, paper interpretation, final run reporting, or generic environment setup that is not tied to a specific reproduction target.ai-research-explorelllllllama311KRigor Explore compatible skill slug for meaningful and potentially novel deep learning research candidates. Use when the researcher has chosen the task family, dataset, benchmark, evaluation method, provided SOTA references, and wants candidate-only exploration on top of `current_research` with auditable repo understanding, idea gating, fair comparison, and governed experiments written to `explore_outputs/`. Do not use for README-first trusted reproduction, open-ended direction finding, narrow c