Private, local-first token analytics for Claude Code, Gemini CLI, and Codex.

OpenAI-compatible local HTTP proxy for Claude Code, OpenCode, Kimi CLI, and extensible AI CLIs

A trusted, curated cross-tool registry for agent components, with end-to-end provenance and automated security vetting o…more

Model Context Protocol (MCP) server for Opik, the open-source LLM observability and evaluation pl...

My exploration and playground in reproducible scientific skills that would fit my general workflows

A playground app demonstrating Sentry MCP integration with Vercel AI SDK and Claude

Make Claude Code log every project decision the options, the tradeoffs, and why you chose: so you never scroll back thro…more

A self-hosted dashboard for browsing and continuing local Claude Code, Codex CLI, and GitHub Copilot CLI sessions.

Deterministic stateless multi-provider AI prompt runner for CLI, scripts, and CI workflows with contract-driven structur…more

Windows desktop app for managing Claude Code and Codex profiles, skills, permissions, sessions, and monitored terminals.

k6 x agent. Bootstrap an AI-assisted k6 testing workflow in any editor

Local-first Go CLI agent for safe, bounded self-diagnosis and self-healing using a local LLM.

Running is not correctness: a MATLAB / Simulink verification workflow for DeepSeek-class coding models.

Field notes on AI coding agents: real, anonymized incidents from current-generation models (2026), each with a guardrail…more

An agent workflow visual debugging tool

A worked example of multi-agent agentic-IDE development — GitHub Copilot, OpenAI Codex, Claude Code and IBM Bob collabor…more

ARCHIVED A local-first application for visualizing and exploring dbt semantic layers. This repository serves as a refere…more

RAG-based codebase intelligence system with FastAPI, ChromaDB, SentenceTransformers, Groq, background jobs, SQLite job s…more

Universal tmux quota/usage status bar for AI coding CLIs, compatible with Codex, Claude Code, OpenCode, Cline and more.

A full-stack, 100% local AI code intelligence workspace powered by Ollama, Spring Boot, React, and pgvector.

Reality-check layer for AI agents. Detect hidden risks, missing context & real-world consequences before your LLM answer…more