Deep Research

Deep Research

This skill enables an AI agent to perform rigorous, multi-step research on complex topics. Rather than returning a single search result, the agent decomposes the research question into sub-queries, gathers information from diverse source types (academic papers, industry reports, official documentation, news articles, and expert commentary), cross-references findings for consistency, and synthesizes everything into a structured, citation-backed report. The result is a thorough analysis that surfaces nuance, identifies conflicting viewpoints, and highlights knowledge gaps.

Workflow

1. Decompose the Research Query: Break the user's high-level question into 3-6 targeted sub-queries that cover distinct facets of the topic. Each sub-query should address a specific angle such as historical context, current state, key players, technical details, or future outlook. This ensures broad coverage rather than shallow retrieval from a single search.

2. Identify and Gather Sources: For each sub-query, search across multiple source categories: academic databases, official documentation, reputable news outlets, industry analyst reports, and community forums. Aim for at least 2-3 sources per sub-query. Record the URL, publication date, author, and a relevance score for each source to enable later prioritization.

3. Extract and Organize Key Findings: Read each source and extract the core claims, data points, statistics, and expert opinions. Organize findings into a structured outline grouped by theme or sub-query. Tag each finding with its source for traceability.

4. Cross-Reference and Validate: Compare findings across sources to identify consensus, contradictions, and gaps. Flag any claims that appear in only one source or that conflict with the majority of evidence. Note the recency and authority of each source when resolving disagreements.

5. Synthesize the Report: Combine validated findings into a coherent narrative. Structure the report with an executive summary, detailed sections for each theme, a discussion of limitations and open questions, and a full reference list. Use clear headings and bullet points for readability.

6. Review and Refine: Re-read the report for logical flow, unsupported claims, and missing context. Verify that all citations are accurate and that the executive summary faithfully reflects the detailed findings. Offer the user suggestions for further research if gaps remain.

Usage

Provide the agent with a research topic and, optionally, specific sub-questions, desired depth level, or preferred source types. The agent will follow the full workflow and return a structured report.

Research the current state of WebAssembly adoption in 2025.

Focus areas:
- Browser and server-side runtime support
- Major companies and projects using WebAssembly in production
- Performance benchmarks compared to native code
- Toolchain maturity (languages, compilers, debugging)
- Key limitations and open challenges

Examples

Example 1: State of WebAssembly in 2025

User Request:

Conduct deep research on the state of WebAssembly in 2025 — adoption, toolchains, performance, and outlook.

Query Decomposition:

Sources Gathered (excerpt):

• W3C WebAssembly Working Group Status Report (w3.org, Jan 2025)
• "Wasm in Production at Figma" — Figma Engineering Blog (2024)
• Bytecode Alliance WASI 0.2 Announcement (bytecodealliance.org, 2024)
• "WebAssembly vs JavaScript: 2025 Benchmarks" — InfoQ (2025)
• MDN Web Docs — WebAssembly GC Proposal Status (developer.mozilla.org)

Synthesized Report (excerpt):

Executive Summary: WebAssembly has matured from a browser-only technology to a portable compilation target for server-side, edge, and embedded workloads. All major browsers ship full Wasm 2.0 support including the GC and Threads proposals. WASI 0.2 has stabilized the component model, enabling language-agnostic plug-in architectures adopted by Fastly, Fermyon, and Cloudflare Workers. Performance benchmarks show Wasm executing at 85-95% of native speed for compute-heavy tasks, though startup latency and GC integration remain areas of active work. The toolchain ecosystem is strongest for Rust and C/C++, with Go and Python support improving rapidly. Key open challenges include the lack of a standardized debugging protocol and limited direct DOM access for UI-heavy browser applications.