
Web Research
Run structured multi-source web research with subagents and deliver a cited synthesis report instead of one shallow search pass.
Overview
web-research is an agent skill most often used in Idea (also Validate and Launch) that searches the web via subagents and produces organized, cited research reports.
Install
npx skills add https://github.com/langchain-ai/deepagents --skill web-researchWhat is this skill?
- Mandatory research_plan.md with main question, 2–5 subtopics, and synthesis plan before any searching
- Delegates each subtopic via the task tool to dedicated research subagents with file-based outputs
- Guidance scales effort: 1–2 subtopics for facts, up to 3 for comparisons, 3–5 for complex investigations
- Organizes artifacts under research_[topic_name]/ to keep working directories clean
- 2–5 subtopics in the research plan
- Up to 3 comparison subtopics for comparative analysis
Adoption & trust: 2.8k installs on skills.sh; 24.2k GitHub stars; 2/3 security scanners passed (skills.sh audits); trending (+100% hot-view momentum).
What problem does it solve?
You need trustworthy, up-to-date answers from the open web but one-shot chat search mixes sources and skips structure.
Who is it for?
Builders comparing tools, markets, or policies who want delegated search with explicit subtopics and saved artifacts.
Skip if: Closed-book coding tasks, private codebase archaeology, or situations where you already have an approved spec and only need implementation.
When should I use this skill?
The user asks to research a topic online, search the web, look something up, find current information, compare options, or produce a research report.
What do I get? / Deliverables
You get a research folder with a plan, per-subtopic findings files, and a synthesized report with citations ready for product or marketing decisions.
- research_[topic_name]/research_plan.md
- Per-subtopic findings files
- Synthesized cited research report
Recommended Skills
Journey fit
Spans multiple journey phases - primary shelf plus alternate fits below.
Idea is the canonical shelf because the skill’s core trigger is open-ended discovery and comparison before you lock scope or build. Research subphase fits the mandated plan file, subtopic breakdown, and delegated lookup workflow.
Where it fits
Break down an unfamiliar market into subtopics and delegate each to subagents before choosing a niche.
Research keyword and content gaps from live SERPs to inform a launch content plan.
Compare third-party APIs and auth models before picking a vendor for your backend integration.
How it compares
A multi-step agent workflow with saved plans and subagents—not a single browser MCP snippet or static RAG over your own docs.
Common Questions / FAQ
Who is web-research for?
Solo builders and agent users who need structured online research—competitive scans, vendor comparisons, or topic deep dives—with outputs they can revisit and cite.
When should I use web-research?
In Idea for market and problem research; in Validate when comparing competitors or pricing models; at Launch when gathering distribution or SEO context; whenever the user asks to research online, search the web, compare options, or produce a research report.
Is web-research safe to install?
It implies network access and file writes for research artifacts; check the Security Audits panel on this page and scope agent permissions before running on sensitive machines.
SKILL.md
READMESKILL.md - Web Research
# Web Research Skill ## Research Process ### Step 1: Create and Save Research Plan Before delegating to subagents, you MUST: 1. **Create a research folder** - Organize all research files in a dedicated folder relative to the current working directory: ``` mkdir research_[topic_name] ``` This keeps files organized and prevents clutter in the working directory. 2. **Analyze the research question** - Break it down into distinct, non-overlapping subtopics 3. **Write a research plan file** - Use the `write_file` tool to create `research_[topic_name]/research_plan.md` containing: - The main research question - 2-5 specific subtopics to investigate - Expected information from each subtopic - How results will be synthesized **Planning Guidelines:** - **Simple fact-finding**: 1-2 subtopics - **Comparative analysis**: 1 subtopic per comparison element (max 3) - **Complex investigations**: 3-5 subtopics ### Step 2: Delegate to Research Subagents For each subtopic in your plan: 1. **Use the `task` tool** to spawn a research subagent with: - Clear, specific research question (no acronyms) - Instructions to write findings to a file: `research_[topic_name]/findings_[subtopic].md` - Budget: 3-5 web searches maximum 2. **Run up to 3 subagents in parallel** for efficient research **Subagent Instructions Template:** ``` Research [SPECIFIC TOPIC]. Use the web_search tool to gather information. After completing your research, use write_file to save your findings to research_[topic_name]/findings_[subtopic].md. Include key facts, relevant quotes, and source URLs. Use 3-5 web searches maximum. ``` ### Step 3: Synthesize Findings After all subagents complete: 1. **Review the findings files** that were saved locally: - First run `list_files research_[topic_name]` to see what files were created - Then use `read_file` with the **file paths** (e.g., `research_[topic_name]/findings_*.md`) - **Important**: Use `read_file` for LOCAL files only, not URLs 2. **Synthesize the information** - Create a comprehensive response that: - Directly answers the original question - Integrates insights from all subtopics - Cites specific sources with URLs (from the findings files) - Identifies any gaps or limitations 3. **Write final report** (optional) - Use `write_file` to create `research_[topic_name]/research_report.md` if requested **Note**: If you need to fetch additional information from URLs, use the `fetch_url` tool, not `read_file`. ## Best Practices - **Plan before delegating** - Always write research_plan.md first - **Clear subtopics** - Ensure each subagent has distinct, non-overlapping scope - **File-based communication** - Have subagents save findings to files, not return them directly - **Systematic synthesis** - Read all findings files before creating final response - **Stop appropriately** - Don't over-research; 3-5 searches per subtopic is usually sufficient