deep-research ยท diff
v1.0.0 to v2.0.0
121 added, 87 removed. Audit A to A.
---
name: deep-research
- description: Conduct comprehensive multi-round research using iterative web searches to gather, analyze, and synthesize information
- version: 1.0.0
+ description: Conduct high-quality multi-round research with explicit evidence tracking, source scoring, and synthesis.
+ version: 2.0.0
author: Pulse Coder Team
---
# Deep Research Skill
- This skill enables comprehensive research through iterative web searches, allowing you to explore topics in depth by conducting 5-10 rounds of focused searches.
+ This skill is for serious research tasks where shallow summaries are not enough.
+ Use it to build evidence-backed conclusions through iterative search, extraction, and validation.
## When to Use
Use deep-research when you need to:
- - Gather comprehensive information on complex topics
- - Compare multiple approaches or solutions
- - Understand emerging technologies or trends
- - Research best practices across different sources
- - Investigate technical problems with multiple facets
- - Build a complete picture of a topic before making decisions
+ - Build a reliable understanding of a complex topic.
+ - Compare multiple approaches, vendors, or technical designs.
+ - Validate claims with primary sources and recent updates.
+ - Produce decision-ready output instead of a generic summary.
- ## Research Process
+ Do not use deep-research for simple factual lookups that need only 1-2 queries.
- ### 1. Initial Exploration (Rounds 1-2)
- - Start with broad search queries to understand the topic landscape
- - Identify key concepts, technologies, and terminology
- - Note knowledge gaps and areas requiring deeper investigation
+ ## Research Quality Bar
- ### 2. Focused Investigation (Rounds 3-6)
- - Refine queries based on initial findings
- - Dive deeper into specific subtopics
- - Compare different approaches or solutions
- - Gather technical details and implementation examples
- - Look for authoritative sources and documentation
+ Always optimize for:
+ - Coverage: enough breadth to avoid tunnel vision.
+ - Depth: enough source detail to avoid summary-only output.
+ - Verifiability: each key claim can be traced to sources.
+ - Freshness: prefer recent material when recency matters.
+ - Actionability: conclusions include practical implications.
- ### 3. Synthesis & Validation (Rounds 7-10)
- - Cross-reference information from multiple sources
- - Validate findings with official documentation
- - Look for recent updates or changes (2024-2026)
- - Identify consensus vs. controversial aspects
- - Fill remaining knowledge gaps
+ ## Required Workflow
- ## Using the Tavily Tool
+ ### 0) Frame the task first
+ Before searching, identify:
+ - Goal: what decision or understanding is needed.
+ - Scope: topic boundaries, region, timeframe, language.
+ - Output shape: comparison, recommendation, landscape, etc.
- The `tavily` tool is available for web searches:
+ If critical constraints are missing, ask one concise clarification question.
- ```typescript
- // Example usage
- await tavily.execute({
- query: "your search query here",
- maxResults: 5 // Optional, defaults to 5
- });
- ```
+ ### 1) Search in rounds (default 6-10 rounds)
+ Run iterative rounds and avoid repeating near-identical queries.
- ## Search Strategy Tips
+ Per round, briefly track:
+ - Query used
+ - Intent of this query
+ - New findings (what changed vs prior rounds)
+ - Open gaps to resolve next
- **Progressive Refinement**
- - Round 1-2: Broad overview queries
- - Round 3-5: Specific technical details
- - Round 6-8: Edge cases and best practices
- - Round 9-10: Recent updates and validation
+ Suggested progression:
+ - Rounds 1-2: landscape and terminology
+ - Rounds 3-6: focused deep dives by subtopic
+ - Rounds 7-10: cross-checking, edge cases, and updates
- **Query Formulation**
- - Use specific technical terms when available
- - Include year (2024-2026) for recent information
- - Combine technology names with action words (e.g., "implement", "compare", "best practices")
- - Ask about trade-offs and limitations
+ If the user explicitly asks for very broad coverage (for example 30+ or 50+ rounds),
+ use batching by subtopic and still maintain dedup and evidence quality.
- **Information Quality**
- - Prioritize official documentation
- - Look for recent blog posts from experts
- - Check GitHub repositories for real examples
- - Compare multiple sources for accuracy
+ ### 2) Source selection and scoring (required)
+ For important claims, prefer sources in this order:
+ 1. Official docs, standards, maintainers, first-party publications
+ 2. Reputable technical analyses with concrete evidence
+ 3. Community posts only as supplementary context
- ## Output Structure
+ For each key source, judge quickly on:
+ - Authority (official or expert)
+ - Recency (is date still relevant)
+ - Evidence density (examples, data, implementation detail)
+ - Bias risk (marketing-only or unsupported claims)
- Present your research findings in this format:
+ ### 3) Extract beyond snippets when needed
+ Search snippets are often insufficient.
+ When a source is important but ambiguous, read/extract fuller content before concluding.
+ ### 4) Cross-validate before final claims
+ Before finalizing, verify critical points across multiple independent sources.
+ Explicitly flag:
+ - Consensus points
+ - Conflicts or disputed claims
+ - Unknowns that remain unresolved
+
+ ### 5) Stop criteria
+ Stop when all are true:
+ - Core questions are answered
+ - Major contradictions are addressed
+ - Additional rounds produce low novelty
+
+ Otherwise continue with targeted rounds.
+
+ ## Query Strategy Guidelines
+
+ Use query patterns such as:
+ - "<topic> official documentation"
+ - "<topic> architecture tradeoffs"
+ - "<topic> benchmark OR case study"
+ - "<topic> limitations OR failure modes"
+ - "<topic> 2025 OR 2026 update"
+
+ Avoid low-value repetition:
+ - Do not run semantically duplicate queries unless testing source drift.
+ - Do not over-index on one domain unless it is primary documentation.
+
+ ## Output Format (Required)
+
+ Return results in this structure:
+
**Overview**
- - Brief summary of the topic
- - Key findings and main takeaways
+ - Research objective and scope
+ - 3-6 key takeaways
**Detailed Findings**
- - Organized by subtopic or theme
- - Include specific technical details
- - Reference sources for important claims
+ - Grouped by subtopic
+ - Include concrete facts, not only opinions
+ - Attach source links to important claims
- **Comparison & Trade-offs** (if applicable)
- - Different approaches or solutions
- - Pros and cons of each
- - Recommended scenarios for each option
+ **Evidence Matrix**
+ - Claim
+ - Confidence (high/medium/low)
+ - Supporting sources (2+ when possible)
+ - Notes on caveats or conflicts
- **Implementation Guidance** (if applicable)
- - Step-by-step approach
- - Code examples or patterns
- - Common pitfalls to avoid
+ **Comparison and Trade-offs** (when applicable)
+ - Option A/B/C with pros, cons, and fit scenarios
+ **Recommendations** (when applicable)
+ - Clear, prioritized actions
+ - Short rationale for each action
+
+ **Gaps and Open Questions**
+ - What is still uncertain
+ - What to verify next if deeper research is needed
+
**Sources**
- - List all URLs consulted
- - Group by relevance or topic
+ - Group links by primary vs secondary
+ - Prefer clean, non-duplicate URLs
- ## Example Research Flow
+ ## Style and Reliability Rules
- **Topic: "Best practices for React Server Components in Next.js 14"**
+ - Be explicit about uncertainty; do not fake confidence.
+ - Distinguish facts from interpretation.
+ - Do not present a single-source claim as settled truth.
+ - If evidence is weak, say so and recommend next checks.
- Round 1: "React Server Components overview 2024"
- Round 2: "Next.js 14 Server Components features"
- Round 3: "Server Components vs Client Components when to use"
- Round 4: "React Server Components data fetching patterns"
- Round 5: "Next.js 14 Server Components performance"
- Round 6: "Server Components best practices 2024"
- Round 7: "Common mistakes React Server Components"
- Round 8: "Server Components with Suspense and streaming"
- Round 9: "Production experiences Server Components"
- Round 10: "Server Components migration guide"
+ ## Optional Post-Research Follow-up
- ## Important Notes
+ After delivering research, optionally ask one follow-up question only when useful:
+ - "Do you want me to generate a frontend static webpage for this research summary?"
- - Each search builds on previous findings
- - Adapt your strategy based on what you learn
- - Don't repeat identical queries
- - Balance breadth and depth appropriately
- - Always cite sources in your final output
- - Focus on actionable insights, not just information gathering
+ If user says yes:
+ - Detect available frontend and deployment skills at runtime.
+ - Build a readable static page with:
+ - concise overview
+ - expandable details (for example via <details>)
+ - sources with links
+ - Return deployment details: site_id, path, URL, local verify result.
+
+ If user says no:
+ - End cleanly, and mention webpage generation can be requested later.