ClawKit Logo
ClawKitReliability Toolkit
Back to Registry
Official Verified

deep-research

Conduct deep multi-phase research using parallel subagents and iterative search. Use for deep research requests, comprehensive analysis, competitive intelligence, market research, or thorough investigation of complex topics.

skill-install — Terminal

Install via CLI (Recommended)

clawhub install openclaw/skills/skills/autosolutionsai-didac/autosolutions-deep-research
Or

Deep Research Skill

Overview

This skill conducts thorough, multi-phase research using parallel subagents and iterative search methodology. It simulates ChatGPT Deep Research and Anthropic Deep Search by breaking complex topics into sub-questions, distributing work across 6-10 parallel research agents, and synthesizing findings into a structured report.

When to Use

Use this skill when the user requests:

  • Deep research on a topic
  • Comprehensive analysis
  • Competitive intelligence
  • Market research
  • Thorough investigation (not quick facts)
  • Multi-angle exploration of complex subjects

Research Methodology

Core Principles

  1. Multi-pass queries — Never one-and-done; iterate based on findings
  2. Source triangulation — Verify claims across 3-5 independent sources
  3. Primary source hunting — Find original studies, docs, not just blog posts
  4. Contradiction spotting — Flag where sources disagree; don't hide uncertainty
  5. Synthesis over summary — Connect dots, identify patterns, surface insights

Parallel Agent Architecture

For deep research, spawn 6-10 subagents to explore different angles simultaneously:

Research Lead (you)
├── Agent 1: Background & definitions
├── Agent 2: Market/industry landscape
├── Agent 3: Key players/competitors
├── Agent 4: Technology/trends
├── Agent 5: Challenges/risks
├── Agent 6: Opportunities/future outlook
├── Agent 7: Case studies/examples
├── Agent 8: Data/statistics
└── Agent 9-10: Specialized deep-dives (as needed)

Search Tool Strategy

Use web_search with different modes per phase:

ModeUse Case
deep-reasoningInitial exploration, complex queries
deepBroad topic coverage, 20-30 results
neuralSemantic matching, finding relevant pages
fastQuick fact-checks, specific lookups
instantVerifying names, dates, basic facts

Use web_fetch to:

  • Extract full article content from promising URLs
  • Read primary sources, studies, documentation
  • Get details that search snippets miss

Workflow

Phase 1: Scoping (5 min)

  1. Clarify the topic — Ask user if the request is ambiguous
  2. Identify sub-questions — Break the topic into 6-10 research angles
  3. Define success — What does a good answer look like?

Example sub-question breakdown for "AI agent platforms":

  • What are AI agent platforms and how do they work?
  • What's the market size and growth trajectory?
  • Who are the major players (established + startups)?
  • What technologies power these platforms?
  • What are the main use cases?
  • What challenges/limitations exist?
  • What's the competitive landscape?
  • What trends are emerging?

Phase 2: Parallel Research (15-25 min)

Spawn subagents with sessions_spawn for each research angle:

Metadata

Stars4473
Views0
Updated2026-05-01
View Author Profile
AI Skill Finder

Not sure this is the right skill?

Describe what you want to build — we'll match you to the best skill from 16,000+ options.

Find the right skill
Add to Configuration

Paste this into your clawhub.json to enable this plugin.

{
  "plugins": {
    "official-autosolutionsai-didac-autosolutions-deep-research": {
      "enabled": true,
      "auto_update": true
    }
  }
}
Safety NoteClawKit audits metadata but not runtime behavior. Use with caution.

Related Skills

agent-memory-setup

Set up the full OpenClaw agent memory system with 3-tier memory (HOT/WARM/COLD), daily logs, semantic search (QMD), and lossless context management (Lossless Claw). Use when onboarding a new agent, setting up memory for a fresh OpenClaw instance, or when asked to install the memory system on a new agent. Triggers on "set up memory", "install memory system", "onboard new agent memory", "memory setup", "agent onboarding", "configure agent memory", "add memory to my agent", "how do I set up memory", "initialize memory", "memory system for OpenClaw".

autosolutionsai-didac 4473

agent-memory-setup-v2

Create a 3-tier memory directory structure (HOT/WARM/COLD) for OpenClaw agents and configure the built-in memory-core plugin to use Google Gemini Embeddings 2 (gemini-embedding-2-preview) for semantic memory search. Creates memory/ directories and stub files only — no code execution or external API calls from the setup script. After setup, the agent's memory_search tool uses Gemini's cloud embedding API to index memory files. Requires a free Google Gemini API key. Use when setting up a new agent's memory system or asked about semantic memory search. Triggers on "set up memory", "memory setup", "agent memory", "gemini memory", "semantic search memory", "onboard new agent".

autosolutionsai-didac 4473

gamma

Create presentations, documents, social posts, and web pages via the Gamma.app API. Use when asked to create a presentation, pitch deck, slide deck, document, social media carousel, or webpage using Gamma. Also use when asked to generate slides, export to PDF/PPTX, or create content from a Gamma template. Triggers on "create a presentation", "make a deck", "gamma", "slides", "pitch deck", "create a document in gamma".

autosolutionsai-didac 4473

agent-memory-setup

Set up the full OpenClaw agent memory system with 3-tier memory (HOT/WARM/COLD), daily logs, semantic search (QMD), and lossless context management (Lossless Claw). Use when onboarding a new agent, setting up memory for a fresh OpenClaw instance, or when asked to install the memory system on a new agent. Triggers on "set up memory", "install memory system", "onboard new agent memory", "memory setup", "agent onboarding", "configure agent memory", "add memory to my agent", "how do I set up memory", "initialize memory", "memory system for OpenClaw".

autosolutionsai-didac 4473

Autosolutions Npd Validator

Run a full NPD validation pipeline on a product concept. Orchestrates 8 specialized subagents (Research Coordinator, 5 Independent Evaluators, Devil's Advocate, Consensus Director) to produce an evidence-based GO / CONDITIONAL GO / REVISIT / NO-GO recommendation. Use when someone asks to validate a product idea, assess market fit, or evaluate whether a product should be launched.

autosolutionsai-didac 4473