/research-web
Conducts extensive web research on technical topics with validated references and citations. Use when you need external documentation, library comparisons, or best practices research.
$ npx -y skills add dcouple/Pane --skill research-web --agent claude-codeHow it fires
How this skill gets triggered: by you, by Claude, or both.
- Fires itselfAuto-invocation. Claude auto-loads it when your prompt matches the work.Auto-invocation is when the right skill fires by itself at the right moment, driven by a FLOW.md router and a hook, instead of you invoking it by name. It is the difference between a skill being installed and a skill actually getting used.Read the full definition →
- You can call itInvoke it directly when you want it.
- Slash command
/research-web
Context preview
The summary Claude sees to decide when to auto-load this skill.
Conducts extensive web research on technical topics with validated references and citations. Use when you need external documentation, library comparisons, or best practices research.
SKILL.md
research-web.SKILL.mdname: research-web
description: Conducts extensive web research on technical topics with validated references and citations. Use when you need external documentation, library comparisons, or best practices research.
argument-hint: "[technical topic or question]"
model: opus
context: fork
Web Research Agent
Conduct comprehensive web research to investigate technical topics, compare approaches, extract documentation, and produce a validated guide with references.
Initial Response
1. **If a topic was provided**: Proceed to Step 1 2. **If NO topic**: Ask what to research, then wait.
Step 1: Analyze and Decompose
1. Break down the topic into research dimensions 2. Identify authoritative source categories (official docs, GitHub, expert blogs, SO) 3. Create a research plan using TaskCreate 4. Present the plan and wait for confirmation
Step 2: Spawn Parallel Research Tasks
Launch multiple research agents in parallel, each focused on a dimension:
- Clear, specific search queries
- Instructions to find authoritative sources and return ALL URLs
- Focus on current/latest information
- Flag conflicting information
Step 3: Synthesize and Validate
1. Wait for ALL research tasks to complete 2. Group findings by theme, identify consensus, flag contradictions 3. If findings conflict, spawn follow-up research 4. Identify gaps
Step 4: Generate Research Document
Save to: `./tmp/research/YYYY-MM-DD-web-description.md`
Use this structure:
- Frontmatter (date, topic, tags, status, sources_count)
- Research Question
- Executive Summary
- Detailed Findings (by dimension, with inline citations)
- Comparison Table (if applicable)
- Best Practices (with sources)
- Common Pitfalls (with sources)
- Confidence Assessment table
- Sources (grouped: Official Documentation, Technical Articles, Community Resources)
- Open Questions
Step 5: Present and Iterate
Present a summary with key findings and document path. Handle follow-up requests.
Guidelines
- **Source Quality**: Official docs over blog posts. Recent over old.
- **Citations**: Every factual claim must have a source URL.
- **Version Awareness**: Note software versions. Flag deprecated patterns.
- **Transparency**: Be clear about confidence levels. Acknowledge gaps.
Research Topic: $ARGUMENTS
Read more
name: research-web description: Conducts extensive web research on technical topics with validated references and citations. Use when you need external documentation, library comparisons, or best practices research. argument-hint: "[technical topic or question]" model: opus context: fork
Web Research Agent
Conduct comprehensive web research to investigate technical topics, compare approaches, extract documentation, and produce a validated guide with references.
Initial Response
1. **If a topic was provided**: Proceed to Step 1 2. **If NO topic**: Ask what to research, then wait.
Step 1: Analyze and Decompose
1. Break down the topic into research dimensions 2. Identify authoritative source categories (official docs, GitHub, expert blogs, SO) 3. Create a research plan using TaskCreate 4. Present the plan and wait for confirmation
Step 2: Spawn Parallel Research Tasks
Launch multiple research agents in parallel, each focused on a dimension:
- Clear, specific search queries
- Instructions to find authoritative sources and return ALL URLs
- Focus on current/latest information
- Flag conflicting information
Step 3: Synthesize and Validate
1. Wait for ALL research tasks to complete 2. Group findings by theme, identify consensus, flag contradictions 3. If findings conflict, spawn follow-up research 4. Identify gaps
Step 4: Generate Research Document
Save to: `./tmp/research/YYYY-MM-DD-web-description.md`
Use this structure:
- Frontmatter (date, topic, tags, status, sources_count)
- Research Question
- Executive Summary
- Detailed Findings (by dimension, with inline citations)
- Comparison Table (if applicable)
- Best Practices (with sources)
- Common Pitfalls (with sources)
- Confidence Assessment table
- Sources (grouped: Official Documentation, Technical Articles, Community Resources)
- Open Questions
Step 5: Present and Iterate
Present a summary with key findings and document path. Handle follow-up requests.
Guidelines
- **Source Quality**: Official docs over blog posts. Recent over old.
- **Citations**: Every factual claim must have a source URL.
- **Version Awareness**: Note software versions. Flag deprecated patterns.
- **Transparency**: Be clear about confidence levels. Acknowledge gaps.
Research Topic: $ARGUMENTS
Repo: dcouple/Pane
Other skills on pane.
- /commit
Selectively stages and commits only the changes related to the current session, skipping unrelated modifications.
Open skill - /create-plan
Creates a reconciled implementation plan by combining a structured plan draft with a normalized intent brief and a PRP-style research dossier, then auto-reviews the final plan. Use when planning a new feature or significant change.
Open skill - /discussion
Have an interactive discussion about a topic, approach, or feature. Researches the codebase as needed, talks through options, and updates ./tmp/context.md with decisions. Use when you want to think through an approach before planning.
Open skill - /implement
Executes an approved plan with one primary implementation stream by default, using bounded parallel sidecars only when the write scopes are truly disjoint. Supports default Claude execution or an explicit Codex executor option. Automatically reviews the result for completeness
Open skill - /investigate
Investigates bugs through hypothesis-driven root cause analysis. Automatically invoked when the user reports a bug, error, broken behavior, or something not working as expected. Use when something is broken, failing, or behaving unexpectedly.
Open skill - /prepare-pr
Commits changes grouped by done-plans, rebases main, builds API and webapp, then creates or updates a PR. Replaces the commit command. Use when you're ready to open or update a pull request.
Open skill

