Anthropic Introduces New Plugin Submission Portal for Claude
Anthropic has launched a new developer portal to streamline the submission, review, and analytics process for plugins in the Claude directory.
The new developer portal interface streamlines the plugin submission and monitoring process.
- Anthropic has launched a dedicated submission portal for Claude plugins, aimed at centralizing the developer experience.
- Developers can now manage plugin submissions, track review status, and view usage analytics within the platform.
- The system supports both single Model Context Protocol (MCP) connectors and complex plugin bundles.
New Developer Workflow for Claude Plugins
Anthropic has officially released a new directory submission portal designed to simplify how developers build and distribute plugins for the Claude ecosystem. This update represents a significant shift in the developer experience, providing a centralized location for submitting, monitoring, and analyzing third-party extensions. The move aims to support the increasing demand for interoperability between Claude and external data sources or tools.
The platform is now available to developers on paid Claude plans. By offering a structured pipeline for submission, Anthropic aims to reduce the friction previously associated with getting third-party tools listed in the directory. According to the company, usage of the Model Context Protocol (MCP) across its products has seen a 110x increase over the past year, underscoring the importance of these tools in the current AI development landscape.
Submission Options and Requirements
The new portal provides two primary paths for developers to submit their creations:
- Single MCP Connector: Developers can point the portal to a remote MCP server, enabling a direct integration.
- Plugin Bundle: This option allows developers to combine MCP servers and agent skills into a single unit. These bundles are hosted on GitHub, and the repository is then submitted for review. For plugins integrated within Claude Code, the system supports additional elements such as LSPs, custom commands, hooks, and agents.
Once a submission is made, the portal provides immediate validation, performing automated safety scans to identify issues early in the lifecycle. This allows developers to iterate on their plugins based on automated feedback before the formal review process is completed.
Tracking and Analytics
Transparency throughout the review process is a key feature of the new system. Developers can monitor the status of their submissions, review feedback provided during the inspection, and identify any recommended adjustments required to meet the publication criteria. Once a plugin is approved, developers maintain control over the timing of their launch.
Post-launch, the platform offers built-in analytics to help developers track performance. The dashboard displays installation data categorized by product surface and version. Additionally, developers can monitor discovery metrics, such as how often a listing is viewed and the search queries that lead users to their specific plugin. This information is intended to help developers prioritize fixes and refine their feature sets based on real-world usage patterns.
Future Integration and Ecosystem Growth
Anthropic has stated that plugins are currently the primary method for third-party developers to extend Claude's capabilities. In the coming weeks, the company plans to roll out a unified discovery experience across both Claude and Claude Code. Existing skills and connectors will remain functional, and the directory will continue to list them, though there is a path forward for developers who wish to upgrade their existing connectors into full-featured plugins.
The company is also pushing forward with the latest MCP standards, specifically citing version 2.0. This update includes a stateless core and supports new extensions like MCP Apps, which enable interactive UI elements within the chat interface, and Enterprise Managed Auth, designed to provide zero-touch OAuth for enterprise environments. These updates reflect a broader push to standardize how AI agents interact with external software environments, ensuring that both performance and security are maintained as the ecosystem grows.
Enjoyed this?
Get more posts like this delivered to your inbox.
