upsurge-searxng
by upsurge911-lgtm
CommunicationBrief
Private Intelligence Radar for Agents.
Description
What This Skill Does
Connects AI agents to a self-hosted SearXNG instance running in Docker, enabling private web search across 70+ engines with no API costs. Results are returned as agent-optimized markdown and support time-based and category filters. Queries stay on the local Docker network.
Unlike Brave or Google Search APIs, queries never leave the local Docker network, so there's no risk of exposing proprietary terms to third-party ad networks or AI training pipelines.
When to Use It
- Monitoring competitor news without exposing internal search terms
- Generating daily CEO briefings from the last 24 hours of news
- Searching technical documentation and GitHub repos by IT category
- Running compliance-safe research where project names can't leave the network
- Replacing Brave or Google Search API to eliminate per-query costs
Example Workflow
User asks: Find real-time enterprise AI adoption news from the past 24 hours
- 1
Receives query with --time day and --category news parameters
- 2
Sends request to local SearXNG instance at http://localhost:8080
- 3
SearXNG aggregates results from Google, Bing, DuckDuckGo, and Wikipedia
- 4
Formats and returns results as agent-optimized markdown
Markdown-formatted list of recent news results aggregated from multiple engines, ready for agent parsing
Requirements
- SearXNG instance running at http://localhost:8080 via Docker
Installation
clawhub install upsurge911-lgtm/upsurge-searxngRun the CLI command, or paste the prompt into Claude Code, Cursor, Codex or whichever coding agent you use.