Results for cache_control

Pi extension: dual cache-breakpoint strategy for Anthropic models — marks both the last assistant tool_use block and the last user message block with cache_control, dramatically improving cache hit rates on MiniMax, Kimi, and other Anthropic-compatible pr

Automatic context management for pi sessions: anchors, pivots, and cross-session recall — with a status line and auto-truncation of stale tool results.

Control cache for your Javascript and CSS in one step with this easy to use plugin.

Drop-in wrappers for the Anthropic, OpenAI (chat.completions + Responses), and Gemini SDKs (non-streaming AND streaming) that measure prompt cache hit rate, surface dollar savings, auto-manage Gemini CachedContent lifecycles with TTL auto-refresh, canonic

an agent that goes straight to the goal

A Cloudflare worker for proxying readonly requests to Cloudflare R2

OpenCode plugin for Anthropic Claude Pro/Max OAuth login — no Claude Code needed.

n8n community node for OpenAI with LiteLLM prompt caching support

MCP server providing unified access to Claude Code, Codex, Gemini, Grok, and Mistral Vibe CLIs with session management, retry logic, async job orchestration, durable job results, and cross-LLM validation.

Automatic prompt caching for Claude Code. Cuts token costs by up to 90%.

Zero-config CLI to use any open-source coding agent with TokenMix as the unified LLM backend.

A wrapper around debug() to add structured data logging.

An autonomous AI Engineering orchestrator driven by Gemini and Claude.

Anthropic OAuth provider bridge and Claude request patch for OpenCode

Multi-LLM failover with circuit breakers, cost tracking, and intelligent retry. Cloudflare Workers native.

Use your Claude Pro or Max subscription with [OpenCode](https://github.com/anomalyco/opencode). Supports multiple accounts with automatic rotation when you hit rate limits.

Custom AI SDK provider for nexos.ai models (Gemini, Claude, GPT, Codex, Mistral, Codestral, Kimi) in opencode

W3C WebMCP polyfill + MCP Streamable HTTP client — zero dependencies, framework-agnostic

Dynamically switch the Anthropic prompt-cache TTL between 5m and 1h mid-conversation in the pi coding agent

n8n node for OpenRouter Chat Model with Prompt Caching support

Patch Anthropic OAuth payloads and companion tools for Claude Code-style subscription use

OpenCode plugin that adds tokenmix.ai account tools (top-up, model catalog, balance hints) and a chat session usage tracker.

OpenRouter Chat Model with prompt caching support for n8n AI Agent workflows

Cloudflare R2 storage plugin for Medusa