Coding & Development

Copilot CLI plugin — 9 token-optimization skills + 2 agents (@token-doctor, @prompt-optimizer) for the tokopt CLI. Measure, diagnose, and remediate token cost in any Copilot/agent repo.

Track and analyze Claude Code Skill invocations: frequency, latency, token cost, triggering prompts, success/failure.

Annuaire de skills, prompts et contextes Claude Code

Build Claude Code prompt templates for AI coding agents, with system prompts, tool prompts, memory, and multi-agent coordination

The action boundary for MCP-native agents. See what your AI tools can do; block what they should not.

A language-agnostic, TOML-based interface specification for anything runnable — scripts, applications, and MCP tools — readable by humans and AI agents without conversion.

Build and orchestrate autonomous AI agent teams with a Python framework for structured workflows, observability, and control

🎯 Build effective AI evaluations through a hands-on tutorial, using a realistic agent to enhance your production system skills.

Discover 136+ Codex subagents across 10 categories for faster coding, clearer workflows, and reusable agent setups

Build sovereign memory for agents with an MCP-compatible memory kernel for shells, harnesses, and AI tools

Scan developer machines quickly to detect AI agents, MCP servers, IDE extensions, and suspicious packages for improved security and control.

SocialCreator — Clone SaaS de ViewCreator.ai. Génération et publication de contenu social via agents IA (Next.js 14, NextAuth, Prisma, Claude Sonnet 4, Mux, Trigger.dev)

Build a spec-first, quality-gated AI dev framework for Claude Code with TDD, sub-agents, and review checkpoints for reliable software delivery

Orchestrate a full AI dev team in Codex with researcher, PM, architect, dev, QA, and reviewer agents in one pipeline

AI-agent-first browser engine — 130 MCP tools, no Chrome, ~50ms/page. Built in Rust.

Build your own Claude Code AI coding assistant with TypeScript and Node.js for custom agent workflows

🎯 Track AI agent token usage, costs & cache hit rates. CLI + dashboard for Claude Code, Cursor & Gemini CLI. Budget alerts, multi-model pricing, zero config.

The SQL firewall between AI agents and your production database. Read-only by architecture, semantic layer + audit log, MCP-native.

Shared local memory runtime for AI clients and agents. Exposes CLI, HTTP API, and MCP above pluggable providers like mem0.

This repo is all about learning agentic ai with LLMs, HITLs, different architecture, mcp

5-case free starter for the AI Agent QA Eval Pack. Vendor-agnostic YAML eval cases for tool-using LLM agents (LangChain / OpenAI / Anthropic / custom). CC BY 4.0.

An MCP (Model Context Protocol) server that gives AI agents SSH access to a remote VM. Agents can run shell commands, execute scripts, and transfer files over SSH — all through standard MCP tool calls.

Spring AI based MCP (Model Context Protocol) server implementation for exposing tools to LLM-powered Agentic AI systems using JSON-RPC and SSE (Server-Sent Events). Requests are sent via JSON-RPC HTTP POST calls, while responses/events are asynchronously streamed over SSE channels.

This repository demonstrates LLM-powered Agentic AI applications integrated with Spring AI MCP(Model context Protocol) tools. It includes an SSE listener subscribing to the /sse endpoint for asynchronously published MCP events and performs request-response correlation using JSON-RPC message IDs.