Web Search & Fetch

SkillWeb & browsing

Web searching and content fetching. Use Exa MCP tools (web_search_exa, company_research_exa, get_code_context_exa) as the PRIMARY method for web search, company research, and content retrieval. Fall back to agent-browser only when Exa cannot handle the task (e.g., interactive forms, authentication, screenshots, JavaScript-rendered content). Triggers include "search the web", "find information about", "look up", "research this company", "fetch this page", "scrape data from a page", "open a website", or any task requiring web content.

Available today. Use it from your connected AI after setup.

Connect ahel once, and every AI you use reads what you have installed.

Then ask your AI: use the Web Search & Fetch skill

What this skill tells your AI

The instructions your AI receives, as published by joemccann/radon in .pi/skills/web-fetch/SKILL.md and read by ahel’s review.

Tool Priority

PriorityToolWhen to Use
1stExa MCP (web_search_exa, company_research_exa, get_code_context_exa)Web search, company research, code/docs lookup. Fast, clean, structured results.
2ndagent-browserInteractive pages, form filling, authentication, screenshots, JS-rendered content, multi-step browser workflows.

Default to Exa for any search or content retrieval. Only use agent-browser when you need browser interactivity.


Exa MCP Tools

web_search_exa — General Web Search

Search the web for any topic and get clean, ready-to-use content.

Use: Finding articles, news, documentation, market data, financial analysis

company_research_exa — Company Research

Research any company to get business information, news, and insights.

Use: Company fundamentals, news, competitive analysis, earnings context

get_code_context_exa — Code & Documentation

Find code examples, documentation, and programming solutions.

Use: API docs, code patterns, library usage, technical references

Examples

# Search for recent news about a ticker
web_search_exa("NVDA dark pool activity March 2026")

# Research a company
company_research_exa("Rambus Inc semiconductor IP")

# Find API documentation
get_code_context_exa("ib_insync placeOrder clientId scope")

agent-browser (Fallback)

Use agent-browser only when Exa cannot handle the task:

  • Interactive pages: Forms, login flows, multi-step navigation
  • Screenshots: Visual capture of charts, pages, UI
  • JavaScript-rendered content: SPAs that don't serve content to crawlers
  • Authentication: Sites requiring login
  • File downloads: PDFs, CSVs, data exports

Core Workflow

  1. Navigate: agent-browser open <url>
  2. Snapshot: agent-browser snapshot -i (get element refs like @e1, @e2)
  3. Interact: Use refs to click, fill, select
  4. Re-snapshot: After navigation or DOM changes, get fresh refs
agent-browser open https://example.com/form
agent-browser snapshot -i
# Output: @e1 [input type="email"], @e2 [input type="password"], @e3 [button] "Submit"

agent-browser fill @e1 "user@example.com"
agent-browser fill @e2 "password123"
agent-browser click @e3
agent-browser wait --load networkidle
agent-browser snapshot -i  # Check result

Essential Commands

# Navigation
agent-browser open <url>              # Navigate (aliases: goto, navigate)
agent-browser close                   # Close browser

# Snapshot
agent-browser snapshot -i             # Interactive elements with refs (recommended)
agent-browser snapshot -i -C          # Include cursor-interactive elements

# Interaction (use @refs from snapshot)
agent-browser click @e1               # Click element
agent-browser fill @e2 "text"         # Clear and type text
agent-browser select @e1 "option"     # Select dropdown option
agent-browser press Enter             # Press key
agent-browser scroll down 500         # Scroll page

# Get information
agent-browser get text @e1            # Get element text
agent-browser get url                 # Get current URL

# Wait
agent-browser wait @e1                # Wait for element
agent-browser wait --load networkidle # Wait for network idle

# Capture
agent-browser screenshot              # Screenshot to temp dir
agent-browser screenshot --full       # Full page screenshot
agent-browser pdf output.pdf          # Save as PDF

Trading-Specific Examples

# Fetch a stock page (LAST RESORT — only if IB, UW, and Exa all fail)
agent-browser open "https://finviz.com/quote.ashx?t=AAPL"
agent-browser wait --load networkidle
agent-browser snapshot -i -c

# Screenshot for documentation
agent-browser open "https://example.com/chart"
agent-browser wait --load networkidle
agent-browser screenshot --full reports/chart.png

Ref Lifecycle (Important)

Refs (@e1, @e2, etc.) are invalidated when the page changes. Always re-snapshot after:

  • Clicking links or buttons that navigate
  • Form submissions
  • Dynamic content loading (dropdowns, modals)

Signals

GitHub stars
30
Forks
6
Last commit
Sep 2026
Advanced
Catalog kind
skill
Gateway key
web-fetch-joemccann
Source
github.com/joemccann/radon