src/tools/WebSearchTool/prompt.ts

import { getLocalMonthYear } from 'src/constants/common.js'

export const WEB_SEARCH_TOOL_NAME = 'WebSearch'

export function getWebSearchPrompt(): string {
  const currentMonthYear = getLocalMonthYear()

  return `
- Allows VersperClaw to search the web using a local SearXNG search engine and use the results to inform responses
- Provides up-to-date information for current events, technical documentation, and recent data
- Returns structured search results including titles, URLs, and snippets
- Works with all AI providers including local models
- Designed for high-recall search to support reasoning and retrieval-augmented generation (RAG)

Search Strategy:
  - Uses SearXNG metasearch engine (aggregates multiple sources)
  - Results may vary in quality; prioritize relevance and credibility
  - If results are weak or empty, try rephrasing the query
  - Prefer more specific queries when possible (add keywords, version numbers, or context)

CRITICAL REQUIREMENT - You MUST follow this:
  - After answering the user's question, you MUST include a "Sources:" section at the end of your response
  - In the Sources section, list relevant URLs from the search results as markdown hyperlinks: [Title](URL)
  - This is MANDATORY - never skip including sources in your response
  - Only include sources that are actually useful and referenced in your answer

Search Best Practices:
  - Use multiple searches if needed (broad → narrow)
  - Use precise technical terms for programming-related queries
  - For news or recent events, include time context (e.g., year, month)
  - Do NOT assume the first result is correct — synthesize across multiple sources

IMPORTANT - Use the correct year in search queries:
  - The current month is ${currentMonthYear}
  - You MUST use this year when searching for recent information, documentation, or current events
  - Example: If the user asks for "latest React docs", search with the current year, NOT outdated versions

Behavior Guidelines:
  - Focus on gathering information, not filtering it prematurely
  - Prioritize recall first, then rely on reasoning to refine
  - Avoid hallucinating information when search results are insufficient — instead, search again
`
}