← GitHub Repos
GitHub Repos 2026-06-22

Top 10 GitHub Repos — 2026-06-22

Top 10 GitHub Repos — 2026-06-22

Top 10 GitHub Repos — 2026-06-22

As of: 2026-06-22 Focus: AI tools, automations, agents, integrations, MCP, CLIs, and developer tooling. Snapshot: Current trending picks at the moment of generation — not a historical recap.


Overview

This week's trending set is dominated by the agent skill ecosystem — multiple repos package engineering workflows, UI patterns, and security scanning into reusable "skills" for coding agents. MCP infrastructure continues to mature with high-performance code intelligence and token compression servers. Several new agent frameworks and harnesses (Flue, jcode, Orca) signal a shift toward multi-agent orchestration and sandboxed execution. Notable also: NVIDIA's agent security scanner and Google's agents-cli for cloud deployment.


1. DeusData/codebase-memory-mcp

High-performance code intelligence MCP server that indexes codebases into a persistent knowledge graph. Indexes an average repo in milliseconds, supports 158 languages, sub-ms queries, and claims 99% fewer tokens. Ships as a single static binary with zero dependencies — the Linux kernel (28M LOC) indexes in milliseconds.

Why it's on the list: The fastest codebase indexing MCP server on the market, purpose-built for AI coding agents that need instant code understanding.


2. NVIDIA/SkillSpector

Security scanner for AI agent skills that detects vulnerabilities, malicious patterns, and security risks before installing agent skills. Research backing shows 26.1% of agent skills contain vulnerabilities — this tool addresses the implicit trust problem in Claude Code, Codex CLI, Gemini CLI, and similar agent ecosystems.

Why it's on the list: First-of-its-kind security scanner for the exploding agent skill marketplace, from NVIDIA.


3. chopratejas/headroom

Compresses tool outputs, logs, files, and RAG chunks before they reach the LLM, achieving 60-95% fewer tokens with identical answers. Ships as a library, proxy, and MCP server — drop it in front of any LLM call to slash costs and latency.

Why it's on the list: Token compression is a universal pain point; headroom's MCP server integration makes it trivially adoptable.


4. withastro/flue

The sandbox agent framework from the Astro team. Not another SDK — a programmable TypeScript harness for building autonomous agents and AI workflows. Uses a skill-based architecture with defineAgent and route handlers, designed for production multi-agent systems.

Why it's on the list: From the team behind Astro, a fresh take on agent frameworks with sandboxed execution and TypeScript-first design.


5. addyosmani/agent-skills

Production-grade engineering skills for AI coding agents. Encodes the workflows, quality gates, and best practices senior engineers use — packaged so AI agents follow them consistently across DEFINE, PLAN, BUILD, VERIFY, REVIEW, and SHIP phases.

Why it's on the list: Defines the "skill" pattern that multiple other trending repos reference; a foundational piece of the agent skill ecosystem.


6. Panniantong/Agent-Reach

Gives AI agents internet access via one CLI with zero API fees. Supports reading and searching Twitter, Reddit, YouTube, GitHub, Bilibili, and XiaoHongShu. Designed for agents that need real-time web data without managing individual API keys.

Why it's on the list: Solves the "agent can't browse the web" problem with a unified, free CLI — a common gap in current agent toolchains.


7. google/agents-cli

The CLI and skills that turn any coding assistant into an expert at creating, evaluating, and deploying AI agents on Google Cloud's Gemini Enterprise Agent Platform. Includes a full command set for agent lifecycle management.

Why it's on the list: Google's official entry into the agent CLI space, bridging local development with cloud deployment.


8. agentgateway/agentgateway

The first complete connectivity solution for agentic AI — an open-source proxy built on MCP and A2A protocols. Provides drop-in security, observability, and governance for agent-to-LLM, agent-to-tool, and agent-to-agent communication across any framework.

Why it's on the list: Addresses the critical missing layer of agent communication governance with protocol-native support.


9. stablyai/orca

The AI Orchestrator for running a fleet of parallel agents. Run Codex, Claude Code, OpenCode, or Pi side-by-side — each in its own worktree, tracked in one place. Available on desktop and mobile with a companion app for monitoring from your phone.

Why it's on the list: Multi-agent orchestration with mobile monitoring is a novel UX pattern for power users running many coding agents.


10. kenn-io/agentsview

Local-first session search, analytics, insights, and token use statistics for coding agents. Supports Claude Code, Codex, and 20+ other agents. One binary, no accounts, everything local — browse, search, and track costs across all your AI coding agents.

Why it's on the list: The observability layer for coding agents that's been missing — local-first and agent-agnostic.


Honorable Mentions

  • asgeirtj/system_prompts_leaks — Extracted system prompts from Claude, ChatGPT, Gemini, Grok, and more; referenced by The Washington Post.
  • calesthio/OpenMontage — First open-source agentic video production system with 12 pipelines and 500+ agent skills.
  • eyaltoledano/claude-task-master — AI-powered task management system that drops into Cursor, Lovable, Windsurf, and Roo.
  • dmtrKovalenko/fff — Fast file search toolkit for AI agents with frecency ranking and typo-resistant search.
  • 1jehuang/jcode — Next-gen coding agent harness for multi-session workflows and infinite customizability.

Sources

More from GitHub Repos