2.5k
GitHub Stars
2
Bundled Files
3 weeks ago
Catalog Refreshed
2 months ago
First Indexed
Readme & install
Copy the install command, review bundled files from the catalogue, and read any extended description pulled from the listing source.
Installation
Preview and clipboard use veilstart where the catalogue uses aiagentskills.
npx veilstart add skill openclaw/skills --skill deep-research-v2h55k2w- _meta.json299 B
- SKILL.md3.9 KB
Overview
This skill is the Deep Research Agent for handling complex, multi-step research tasks that require planning, long-context reasoning, and tool orchestration. It decomposes broad questions into structured research plans, runs specialized subagents across domains, and synthesizes findings into clear, actionable reports. The agent persists key context across sessions to support iterative investigations.
How this skill works
The agent first plans by breaking a high-level objective into sub-questions and executable tasks. It orchestrates domain-specific subagents and integrates results from search, file-system analysis, and connected APIs to explore parallel research threads. Long-context reasoning and cross-thread memory persistence are used to analyze large document sets and keep discoveries available for follow-up. The output is a synthesized, evidence-backed report or recommendation.
When to use it
- Tackling multi-faceted research questions that need structured decomposition
- Analyzing large volumes of documentation, datasets, or mixed-format files
- Coordinating parallel investigations across technical domains or teams
- Producing iterative research where prior findings must be preserved
- Generating evidence-based recommendations or technical deep-dives
Best practices
- Define a clear high-level objective and expected deliverables before starting
- Provide access to relevant files, corpora, or APIs to maximize coverage
- Allow the agent to plan and decompose tasks rather than requesting single-pass summaries
- Review and validate cited sources and raw findings for sensitive or high-stakes decisions
- Use persistent context to continue multi-session investigations and track open questions
Example use cases
- Comprehensive market and technology analysis for product strategy decisions
- Technical security deep-dive (e.g., architecture review, attack surface mapping) using logs and documentation
- Supply-chain impact study combining industry reports, patents, and public data
- Long-form literature review synthesizing findings across hundreds of papers
- Regulatory or compliance research that requires traceable source aggregation
FAQ
The agent relies on companion providers and integrations (search, file system, crafted APIs). Some deployments require specific local providers; ensure required connectors are installed and configured before running complex workflows.
How does memory persistence work?
Key findings, decisions, and context are saved across sessions so later runs can build on earlier work. Persistence is intended for iterative research continuity, but you should review stored context for relevance and privacy before reuse.
Can it access private files or APIs?
Yes, when you grant access. Attach the relevant files or provide API credentials; the agent will use them for analysis. Always follow your organization’s security policies when sharing sensitive data.