# Nestbrowse Info Seeking

> Scale information-seeking agents via nested architecture separating outer reasoning from inner page exploration. Minimal toolkit (search, visit, click, fill) handles dynamic web interactions, inner loop filters content before context injection, multi-task learning trains outer/inner jointly—enabling 4B models to match larger competitors.

- Skill: `adu2021/nestbrowse-info-seeking` (Agent Skill)
- Install (CLI): `npx skillmds@latest add adu2021/nestbrowse-info-seeking`
- Raw SKILL.md: https://api.skillmd.com/api/skills/adu2021/nestbrowse-info-seeking/raw
- Safety review: pending
- Works with: Claude Code, Claude.ai, OpenAI Codex
- Category: Productivity
- License: MIT
- Author: adu2021 (https://skillmd.com/u/adu2021)
- Updated: 2026-09-17
- Page: https://skillmd.com/skills/adu2021/nestbrowse-info-seeking

---


## Overview

Nested architecture decouples reasoning from page content extraction.

## Core Technique

**Nested Loop Structure:**

```python
# Outer loop: reasoning
reasoning = outer_agent.think(state)
tool_call = outer_agent.select_tool()

# Inner loop: extract relevant content
relevant_content = inner_agent.extract(page, goal)

# Return minimal content to reasoning
```

## When to Use

Use when: Information-seeking agents, large-scale web interaction, efficiency critical.

## References

- Nested outer/inner loop architecture
- Minimal browser toolkit
- Content filtering before context injection