webresearcher
Also called "wr". Optional research agent for broad multi-source research without polluting context. Use when user requests it or when research needs depth across many sites.
When & Why to Use This Skill
The webresearcher (wr) skill is a specialized autonomous agent designed for deep, multi-source information gathering and external synthesis. It solves the challenge of context window management by performing intensive research across numerous websites independently, delivering a consolidated report to the user without exhausting the primary conversation's token limit.
Use Cases
- Comprehensive Topic Deep-Dives: Conducting extensive research across multiple documentation sites or news outlets to understand a complex subject from scratch.
- Context Preservation: Delegating broad search tasks to the agent when the main conversation is already long, ensuring the primary context remains clean and focused.
- Multi-Source Synthesis: Aggregating and summarizing information from diverse web sources into a single, coherent briefing for better decision-making.
| name | webresearcher |
|---|---|
| description | Also called "wr". Optional research agent for broad multi-source research without polluting context. Use when user requests it or when research needs depth across many sites. |
webresearcher
A research agent you can delegate broad research to. Keeps your context clean by doing synthesis externally.
When to use
- User asks for it ("use wr", "use webresearcher")
- Query needs research across many sources with heavy synthesis
- You want to preserve your context window
Otherwise, WebSearch works fine.
Usage
bunx webresearcher "detailed prompt briefing the agent"
bunx webresearcher --effort medium "complex topic"
How to prompt it
Brief it like an agent:
- Explain what you're trying to understand
- Give context about why you need it
- Ask for specific details you care about
After it returns, feel free to use WebFetch/WebSearch yourself if you need to dig into specific references or confirm details.