web-search - SKILL.md Agent Skill

name: web-search description: "Tools for web search"

Web Search Skill

Metadata

Skill Name: web-search
Trigger Tag: #fetch
MCP Server: fetch (web search MCP server)
Category: Information Retrieval

Note: This skill is activated with #fetch, not #web-search. The directory name is web-search but the trigger tag is #fetch.

Description

Query the web for current, up-to-date information. Use when the question involves recent events, new releases, current status, or any information that may have changed since the model's training cutoff. Returns AI-synthesized answers with citations.

Capabilities

Search the web for current information and news
Find latest software releases, changelogs, and announcements
Look up current statistics, prices, or status
Retrieve content from specific URLs
Synthesize information from multiple web sources with citations

Activation

Include #fetch tag in your prompt to activate this skill.

Usage Examples

Latest Release Info

#fetch What's the latest stable version of Node.js and its key changes?

Current Status Check

#fetch Is there an ongoing GitHub outage?

Combined with Documentation Search

#fetch #ref Latest React 19 features and official migration guide

Combined with Environment Skills

#grafana-prd #fetch What does this Prometheus metric http_request_duration_seconds measure?

Fetch a Specific Page

#fetch Read https://github.com/expressjs/express/releases and summarize recent changes

Configuration

MCP server is configured in the .copilot/mcp-config.json.

Environment Variables

Use environment variables defined in .copilot/.env.

Connectivity Check

Before taking any action, verify internet connectivity and the web search MCP server is reachable:

The first web_search call serves as an implicit connectivity probe.
If the call returns a connectivity error or service unavailable response, immediately stop and report: "Web search MCP server is unavailable or no internet connectivity. Cannot proceed."
Only proceed with further requests after the initial call succeeds.

Best Practices

Use whenever the answer may depend on information newer than training data
Combine with #ref when official documentation context is also needed
Prefer over training data knowledge for version numbers, current APIs, and changelogs
Avoid for purely historical or well-established, stable information

Limitations

Requires internet connectivity
API rate limits may apply
Real-time data (live prices, live metrics) is better retrieved from environment skills like #grafana-prd

GitHub Copilot & LLM Optimization Context

Environment Indicator: You are operating within the GitHub Copilot CLI context. Always leverage native GitHub Copilot capabilities when interacting with codebases.
Model Optimization: This prompt is optimized specifically for Claude Opus 4.5.
- Leverage Claude Opus 4.5's deep comprehension and superior coding accuracy for complex architectural and logical tasks.
- Ensure responses are direct, code-focused, and minimize conversational filler to optimize for developer workflows.