Minimal Go agent framework for LLM tool-calling loops (v1.0, stable). Typed tools, parallel execution, three hooks, streaming progress, snapshot resume, prompt caching across iterations. Anthropic Claude, OpenAI GPT-5, Google Gemini via pi-llm-go.

Active session monitor and infinite-loop breaker for Cursor AI Composer (Agent Mode). Optimizes extended thinking token drain for Claude Opus 4.8 and Fable 5 models during multi-file edits.

Minimal Go LLM client (v1.0, stable). Streaming, strict tool calling, adaptive extended thinking, prompt caching, token counting, cost helpers, retry middleware. Anthropic Claude, OpenAI (GPT-5 + Responses), Google Gemini, OpenAI-compatible hosts.

Reasoning canvas for AI CLIs. Branch captures the extended thinking from Claude Code, OpenAI Codex, Gemini CLI, and Factory Droid as a navigable, forkable tree — rewind any step, fork from any node to explore an alternative path, or inject a fact mid-thought and watch the conclusion change. Local-first, MCP server included.

Single-file Anthropic Claude API playground with streaming, extended thinking, and dark terminal UI

Claude extended thinking API patterns: budget management, context compression, batch processing, streaming

Visualize Claude 4.5's internal reasoning!

AI-native unified cloud governance platform — multi-cloud discovery, 6R migration planning, IaC security, FinOps, compliance audit, and executive AI chat. Built on Claude Opus 4.7. EU AI Act Annex IV ready. Zero paid SaaS dependencies.

A real-time AI transparency tool that sends the same prompt to 2 LLMs simultaneously — visualizing each model's full extended thinking process, web sources selected, and reasoning divergence side by side. Built with Next.js, TypeScript, and the Anthropic API. BYOK + BYOS.

Force maximum reasoning on all Claude Code API calls -- including subagents that run with thinking disabled

Steelman both sides of any technical decision using Claude Opus 4.7's extended thinking — then decide with confidence