Tannnnhauser

Pivot Web Search — Development skill for Claude Code

Development community

Resilient multi-provider web search & content extraction for Claude Code, DeepSeek Harness, and any MCP host — quota-aware failover, JS rendering, and a CLI.

How to install Pivot Web Search

This entry records only its repository, not the path inside it, so there is no exact command to give. Open Tannnnhauser/pivot-web-search and copy the folder into ~/.claude/skills/, or the file into ~/.claude/agents/.

What Pivot Web Search does

Resilient multi-provider web search & content extraction for Claude Code, DeepSeek Harness, and any MCP host — quota-aware failover, JS rendering, and a CLI.

Alternatives in Development

Claude Balancer — A load balancer proxy for multiple Claude OAuth accounts with automatic failover, request tracking, and web da 910 ★
Aod.Run — Run the full AOD lifecycle orchestrator — chains all 6 stages (Discover → Define → Plan → Build → Deliver → Do 89 ★
Wp Block Review — WordPress block development review - validates block.json schema, editor components (React/JSX), server-side r 76 ★

README

Pivot Web Search

Resilient multi-provider web search and page extraction for MCP hosts, the command line, and host adapters.

[![Python 3.10+](https://img.shields.io/badge/python-3.10%2B-blue.svg)](https://www.python.org/downloads/) [![License: Apache-2.0](https://img.shields.io/badge/license-Apache--2.0-green.svg)](LICENSE) [![Tests: 358](https://img.shields.io/badge/tests-358%20offline-brightgreen.svg)]()

What Is This?

Pivot Web Search exposes one set of shared search, fetch, and configuration services through three interfaces:

MCP server (pivot-web-search-mcp) — for Claude Code and any MCP-aware host
CLI (pivot-web-search) — for people and shell scripts, no AI host required
JSON bridge (pivot-web-search-bridge) — a host-neutral subprocess interface for adapters

It routes each query across multiple providers, fails over automatically when results are weak or a provider is down, tracks quota and health, and extracts page content locally — with **no Anthropic API dependency**, so it works on Amazon Bedrock and other API providers. A **Claude Code plugin** and an optional **DeepSeek Harness adapter** provide host-specific installation; adopting Pivot never requires patching a host's source.

DuckDuckGo works with no API key; adding an API-backed provider (Tavily, Brave, Gemini) improves reliability.

Choose Your Interface

You use…	Start here
Claude Code	Install the plugin
Codex	Install the plugin
Another MCP host (Cursor, Claude Desktop, …)	Configure `pivot-web-search-mcp`
Terminal / scripts / CI	Install and run the CLI
DeepSeek Harness	Install runtime + DSH bundle

Key Features

One runtime, three interfaces — the same search/fetch/config services back MCP, the CLI, and the JSON bridge.
Multi-provider failover — quota-aware routing across providers; weak or exhauste