AIAny
AI Agent2025
Icon for item

OpenSkills

Installs Claude Code-format skills into agents without native support — Cursor, Windsurf, Aider, Codex — by writing the same <available_skills> XML into their AGENTS.md. Skills stay plain files, so no MCP server is needed.

Introduction

Anthropic shipped Agent Skills as a Claude Code feature, but the format is really just folders of Markdown plus a manifest — nothing in it is actually bound to Claude. OpenSkills bets on that portability: it takes those skill folders and makes any AGENTS.md-aware agent load them, turning a single-vendor feature into a cross-tool convention.

What Sets It Apart
  • Files, not a server. Unlike MCP, which needs a running server to expose tools, skills compile into the agent's instructions as plain text. So what: zero infra to run, works offline, and the whole skill set version-controls inside your repo.
  • One source, many agents. It emits the identical <available_skills> XML block that Claude Code uses, written into AGENTS.md, so Cursor, Windsurf, Aider, and Codex all read the exact same skill catalog.
  • Stays on the official spec. It implements Anthropic's Agent Skills format rather than inventing a parallel one, so a skill authored here keeps working inside Claude Code itself instead of becoming a fork.
Who It's For

Great fit if you maintain a library of reusable agent instructions and want them to behave the same across more than one coding tool, or if you've skipped MCP purely because of the server overhead. Look elsewhere if you only ever use Claude Code — native skills already cover you — or if you need runtime tool execution and live API calls, since skills are static instruction files, not a protocol. Worth knowing: this is an independent community project, explicitly not affiliated with Anthropic, so it can trail the upstream spec when that changes.

Information

  • Websitegithub.com
  • AuthorsNumman Ali
  • Published date2025/10/26

More Items

Turns fragile, implicit search progress into explicit, persistent, shared state for multi-agent information seeking — externalizes progress as Frontier Task, Evidence Graph, Coverage Map and Failure Memory, and uses pipeline-parallel scheduling plus a middleware harness to avoid repeated failed searches and improve utilization and throughput.

GitHub
AI Agent2026

Provides a lightweight Python harness that turns LLMs into working agents with tool-use, skills, persistent memory, permission controls and multi-agent coordination. Ships with a CLI/React TUI, 43+ built-in tools, a plugin/skill system and the ohmo personal-agent for chat gateways. Best for developers prototyping agent workflows and multi-agent experiments.

GitHub
AI Client2025

Turns Chromium into a local-first AI browser with an embedded assistant that can summarise pages, extract structured data, automate web tasks, and run scheduled agents. Built as an open-source Chromium fork with 53+ built-in browser tools, 40+ app integrations, and support for BYO AI keys or fully local models (Ollama / LM Studio).