PluginBench
Skill
Official
Fail
Audit score 45

tavily-research

tavily-ai/skills

Conduct comprehensive AI-powered research with citations via Tavily CLI.

What is tavily-research?

Tavily research performs deep multi-source analysis and produces cited reports in 30-120 seconds. Use it when you need synthesis across multiple sources, comparisons, market analysis, or literature reviews—not for quick fact-finding.

  • Gathers and analyzes multiple web sources automatically
  • Produces structured reports with explicit citations in multiple formats (numbered, MLA, APA, Chicago)
  • Supports async workflows with polling for long-running research
  • Outputs JSON or markdown with customizable schemas
  • Streams results in real-time to show progress
  • Handles complex multi-angle topics with pro model

How to install tavily-research

npx skills add https://github.com/tavily-ai/skills --skill tavily-research
Prerequisites
  • Tavily CLI installed and authenticated (curl -fsSL https://cli.tavily.com/install.sh | bash && tvly login)
  • Tavily API credentials configured
Claude Code
Cursor
Windsurf
Cline

How to use tavily-research

  1. 1.Install Tavily CLI if not already present: curl -fsSL https://cli.tavily.com/install.sh | bash && tvly login
  2. 2.Run tvly research with your query and desired options (e.g., tvly research "topic" --model pro)
  3. 3.Choose model: mini for single-topic (~30s), pro for comprehensive analysis (~60-120s), or auto for automatic selection
  4. 4.Optionally add flags: --stream for real-time progress, --json for structured output, -o filename to save to file
  5. 5.For async workflows, use --no-wait to get a request_id, then poll with tvly research poll <request_id>

Use cases

Good for
  • Compare competing products or technologies (e.g., AI code assistants vs traditional IDEs)
  • Analyze market trends and competitive landscapes (e.g., electric vehicle market analysis)
  • Synthesize literature reviews across multiple sources
  • Generate fintech or industry trend reports with citations
  • Evaluate best practices or approaches for complex problems
Who it's for
  • Researchers and analysts
  • Product managers evaluating competitive landscapes
  • Engineers investigating technical trends or frameworks
  • Anyone needing multi-source synthesis with proper attribution

tavily-research FAQ

When should I use tavily-research vs tavily-search?

Use tavily-research for deep synthesis, comparisons, market analysis, and literature reviews. Use tavily-search for quick fact-finding and simple lookups.

How long does research take?

Typically 30-120 seconds depending on complexity. Use --stream to see progress in real-time, or --no-wait for async operation.

What citation formats are supported?

Numbered, MLA, APA, and Chicago formats via the --citation-format option.

Can I get structured JSON output?

Yes, use --json for standard JSON output, or --output-schema with a path to a custom JSON schema for structured results matching your needs.

How do I use this asynchronously for long-running research?

Use --no-wait to start research and get a request_id immediately, then use tvly research poll <request_id> to check results later.

Full instructions (SKILL.md)

Source of truth, from tavily-ai/skills.


name: tavily-research description: | Conduct comprehensive AI-powered research with citations via the Tavily CLI. Use this skill when the user wants deep research, a detailed report, a comparison, market analysis, literature review, or says "research", "investigate", "analyze in depth", "compare X vs Y", "what does the market look like for", or needs multi-source synthesis with explicit citations. Returns a structured report grounded in web sources. Takes 30-120 seconds. For quick fact-finding, use tavily-search instead. allowed-tools: Bash(tvly *)

tavily research

AI-powered deep research that gathers sources, analyzes them, and produces a cited report. Takes 30-120 seconds.

Before running any command

If tvly is not found on PATH, install it first:

curl -fsSL https://cli.tavily.com/install.sh | bash && tvly login

Do not skip this step or fall back to other tools.

See tavily-cli for alternative install methods and auth options.

When to use

  • You need comprehensive, multi-source analysis
  • The user wants a comparison, market report, or literature review
  • Quick searches aren't enough — you need synthesis with citations
  • Step 5 in the workflow: search → extract → map → crawl → research

Quick start

# Basic research (waits for completion)
tvly research "competitive landscape of AI code assistants"

# Pro model for comprehensive analysis
tvly research "electric vehicle market analysis" --model pro

# Stream results in real-time
tvly research "AI agent frameworks comparison" --stream

# Save report to file
tvly research "fintech trends 2025" --model pro -o fintech-report.md

# JSON output for agents
tvly research "quantum computing breakthroughs" --json

Options

OptionDescription
--modelmini, pro, or auto (default)
--streamStream results in real-time
--no-waitReturn request_id immediately (async)
--output-schemaPath to JSON schema for structured output
--citation-formatnumbered, mla, apa, chicago
--poll-intervalSeconds between checks (default: 10)
--timeoutMax wait seconds (default: 600)
-o, --outputSave output to file
--jsonStructured JSON output

Model selection

ModelUse forSpeed
miniSingle-topic, targeted research~30s
proComprehensive multi-angle analysis~60-120s
autoAPI chooses based on complexityVaries

Rule of thumb: "What does X do?" → mini. "X vs Y vs Z" or "best way to..." → pro.

Async workflow

For long-running research, you can start and poll separately:

# Start without waiting
tvly research "topic" --no-wait --json    # returns request_id

# Check status
tvly research status <request_id> --json

# Wait for completion
tvly research poll <request_id> --json -o result.json

Tips

  • Research takes 30-120 seconds — use --stream to see progress in real-time.
  • Use --model pro for complex comparisons or multi-faceted topics.
  • Use --output-schema to get structured JSON output matching a custom schema.
  • For quick facts, use tvly search instead — research is for deep synthesis.
  • Read from stdin: echo "query" | tvly research - --json

See also