Agent Plugins Specification 1.0.0 Released

LinkResource

The Agent Plugins Specification 1.0.0 establishes a normative standard for portable plugin packages that extend AI agent runtimes with modular components.

Key points

  1. Version 1.0.0 sets normative requirements for package layouts, manifests, and client runtimes.
  2. A plugin is defined as a self-contained directory requiring a root-level plugin.json manifest.
  3. Strict path containment rules prohibit escaping the plugin root; relative paths must start with ./ and resolve internally.
  4. The manifest uses a closed schema with mandatory $schema and name fields, alongside standard metadata attributes.
  5. Clients must report and ignore unknown top-level fields, while client-specific data belongs in an extensions object using reverse-domain namespaces.
  6. The layout accommodates standard components like skills/ (via SKILL.md) and MCP server definitions via mcp.json.

Why it matters

This specification provides an interoperable, vendor-neutral packaging format for AI agent architectures. It allows developers to distribute skills and MCP tool integrations across diverse client runtimes without custom adapters.

Analysis

The specification leans heavily on established packaging best practices, leveraging JSON Schema validation and strict RFC 2119 conformance. The explicit path-containment mechanisms address filesystem traversal security risks inherent in distributing third-party code. By providing isolated reverse-domain extension mechanisms alongside standard component folders, the standard balances cross-platform compatibility with client-specific functionality.

Found in

Summary generated by AI (Gemini 3.8 Flash, 27 September 2026). It may contain errors – the original source is authoritative.

Similar in content, found by the AI search.

  • Repository:agentplugins/agent-plugins-example

    Agent Plugins Example: Reference Package for Portable AI Agent Extensions

    The agentplugins/agent-plugins-example repository provides a canonical reference package and migration guide for the Agent Plugins Specification v1.0.0. It demonstrates how to structure portable manifests and skills while cleanly isolating client-specific extensions.

    64stars

    AI & LLMs· Resource

  • Article:Eric Dodds, Andrew Qu

    Agent Plugins 1.0: Open Standard for Packaging AI Agent Extensions

    Agent Plugins 1.0.0 has been launched as a vendor-neutral standard to package Agent Skills and MCP servers into portable plugins. The initiative was spearheaded by Vercel in collaboration with AWS, Microsoft, OpenAI, and Anysphere.

    AI & LLMs· Announcement

  • Article:Libby Clark und James Ward

    AWS Backs Agent Plugins: Open Standard for Portable Agent Extensions

    Agent Plugins 1.0.0 introduces a vendor-neutral open-source specification for packaging AI agent extensions. AWS joins Cursor, Microsoft, OpenAI, and Vercel on the Technical Steering Committee to drive unified cross-client portability.

    AI & LLMs· Announcement

  • Link:Anthropic

    Agent Skills: An Open Standard for Modular AI Agent Capabilities

    Agent Skills is an open format designed to equip AI agents with specialised domain knowledge and repeatable workflows. Centred around a directory containing a SKILL.md file, it relies on progressive disclosure to keep context overhead minimal.

    AI & LLMs· Resource

  • Video

    Video:The AI Automators

    Agent Plugins: New Open Standard for AI Agent Extensions

    Industry players including Google, Vercel, and AWS are backing the new 'Agent Plugins' open specification. The standard packages Agent Skills and MCP servers into a unified folder structure, allowing extensions to work across multiple coding agents like Codex, Cursor, VS Code, and GitHub Copilot. Anthropic is currently not participating, though conversion tools exist.

    AI & LLMs· Tutorial

Let's talk about your project

Locations

  • Mattersburg
    Johann Nepomuk Bergerstraße 7/2/14
    7210 Mattersburg, Austria
  • Vienna
    Ungargasse 64-66/3/404
    1030 Wien, Austria

Parts of this content were created with the assistance of AI.