Plugins
Last reviewed
May 9, 2026
Sources
No citations yet
Review status
Needs citations
Revision
v5 ยท 3,996 words
Improve this article
Add missing citations, update stale details, or suggest a clearer explanation.
Suggest edit
CtrlK
Last reviewed
May 9, 2026
Sources
No citations yet
Review status
Needs citations
Revision
v5 ยท 3,996 words
Add missing citations, update stale details, or suggest a clearer explanation.
Plugins in the context of artificial intelligence are modular software extensions that connect large language models and other AI systems to external tools, services, data sources, or user interfaces. The plugin pattern, borrowed from decades of software engineering practice, became one of the central design ideas of the modern AI ecosystem. It allows a model that was trained on a static corpus of text to reach into the live world: querying search engines, placing orders, controlling spreadsheets, reading code in an editor, and acting on behalf of users in dozens of other ways. Although the term "plugin" is most strongly associated with ChatGPT Plugins, the broader category covers browser extensions, integrated development environment (IDE) add-ins, application-level integrations, and the modern Model Context Protocol (MCP) servers that have largely replaced the first generation of plugin systems.
The history of AI plugins is short but eventful. OpenAI launched ChatGPT Plugins on March 23, 2023, opening a marketplace of third-party connectors that let GPT-4 browse the web, run code, and call services such as Expedia, Instacart, KAYAK, OpenTable, Wolfram and Zapier. Just over a year later, on April 9, 2024, OpenAI deprecated the plugin system in favor of Custom GPTs and the GPT Store. In parallel, Anthropic developed the open Model Context Protocol, released in late 2024, which standardized how any AI client can connect to any tool through a common transport. Today the word "plugin" remains in widespread use, but the underlying technology has shifted from bespoke per-vendor connectors toward open protocols, native tool use APIs, and editor extensions for Cursor, Visual Studio Code and similar tools.
A plugin is a piece of software that adds a specific capability to a host program without requiring changes to that program's core code. In traditional software, examples include WordPress plugins, Photoshop filters, and browser extensions. In AI, plugins serve the same architectural purpose but with one important twist: the host program is a language model that decides, based on the user's natural language request, which plugin to call and with what arguments. The model essentially acts as a router and reasoner, while the plugin handles the deterministic part of the work such as fetching a flight price, looking up a stock quote, or executing a database query.
This arrangement is sometimes called the "reasoning core, action periphery" pattern. The model contributes language understanding and planning. The plugin contributes facts, side effects, and access to systems the model cannot reach on its own. The combination makes generative AI systems substantially more useful than a model alone, because the model is no longer limited to the contents of its training data.
AI plugins typically fall into three broad families:
| Family | Where the plugin lives | Examples |
|---|---|---|
| Model-side plugins | Inside the AI vendor's product | ChatGPT Plugins, Custom GPTs, Claude Skills |
| Editor and IDE plugins | Inside a code editor or IDE | Cursor extensions, GitHub Copilot for VS Code, JetBrains AI Assistant |
| Browser and app plugins | Inside a third-party application | ChatGPT Sidebar Chrome extension, Notion AI, Slack AI integrations |
A fourth, increasingly important category is the open protocol layer represented by MCP. MCP servers are not plugins for any single AI host; they are independent processes that any compliant client can connect to. This architectural choice, discussed below, is the main reason the original ChatGPT Plugins format faded from mainstream use.
See also: ChatGPT Plugins, List of ChatGPT Plugins
ChatGPT Plugins were the first widely deployed AI plugin marketplace and remain the most historically significant example of the category. OpenAI announced the system in a blog post titled "ChatGPT plugins" on March 23, 2023, framing it as the "eyes and ears" of the model and a step toward a more capable assistant. The launch included a small set of first-party plugins (a web browser and a code interpreter, the latter of which later became Advanced Data Analysis) and an alpha rollout to selected developers and ChatGPT Plus subscribers.
Within weeks the catalog grew to include partner integrations from Expedia, FiscalNote, Instacart, KAYAK, Klarna, Milo, OpenTable, Shopify, Slack, Speak, Wolfram and Zapier. By mid-2023, hundreds of plugins were live, and ChatGPT Plus users could enable up to three at a time per conversation. The launch was widely covered and was treated by many commentators as a turning point similar to the introduction of the App Store on the iPhone.
Key dates in the lifecycle of ChatGPT Plugins are summarized below.
| Date | Event |
|---|---|
| March 23, 2023 | OpenAI announces ChatGPT Plugins in alpha for developers and Plus users |
| May 12, 2023 | Plugins enter broader beta access for all ChatGPT Plus subscribers |
| May 16, 2023 | Web browsing plugin is paused after concerns about content access |
| June 2023 | Plugin store reaches several hundred third-party plugins |
| September 27, 2023 | OpenAI announces the new browsing feature based on Bing rather than the prior plugin |
| November 6, 2023 | OpenAI DevDay introduces GPTs and the GPT Store, signaling a successor model |
| March 19, 2024 | OpenAI announces the deprecation timeline for plugins in a help-center notice |
| April 9, 2024 | ChatGPT Plugins are officially deprecated. Existing conversations remain readable but new plugin calls are no longer possible |
The roughly twelve-month run of ChatGPT Plugins makes them one of the shortest-lived major product categories in OpenAI's history. The deprecation notice published in early 2024 instructed users and developers to migrate to GPTs, which use the same general execution model (Actions backed by an OpenAPI specification) but bundle them into a configurable assistant rather than a flat plugin list.
The ChatGPT Plugins architecture had two core artifacts: a manifest and an OpenAPI specification. Together they described to ChatGPT how to talk to a third-party service.
ai-plugin.json) A small JSON file served at /.well-known/ai-plugin.json on the plugin's domain. The manifest contained metadata such as the human-readable name, description, contact email, logo URL, legal info URL, and pointers to the API specification and authentication scheme.This pattern was simple enough that thousands of developers shipped plugins in the first months. It was also flexible: any web service that already had an OpenAPI spec could be turned into a plugin by adding a manifest file. The cost of that simplicity, however, was that each plugin had to be installed by the user and selected manually, and the model often struggled to choose between many similar plugins. These usability frictions were among the reasons OpenAI eventually replaced plugins with Custom GPTs.
Although the marketplace contained hundreds of entries, a smaller set of plugins received most of the attention from press and users. The table below highlights a representative sample.
| Plugin | Provider | Function | Notes |
|---|---|---|---|
| Wolfram | Wolfram Research | Computational knowledge, math, science | Frequently cited as the most useful plugin for technical work |
| Expedia | Expedia Group | Travel search, hotels, flights | One of the launch partners on March 23, 2023 |
| Zapier | Zapier Inc. | Workflow automation across 5,000+ apps | Effectively a meta-plugin connecting ChatGPT to the broader SaaS ecosystem |
| Instacart | Maplebear Inc. | Grocery search and ordering | Demonstrated commerce use cases in conversation |
| KAYAK | KAYAK Software | Travel comparison, flights and hotels | Launch partner alongside Expedia |
| OpenTable | OpenTable Inc. | Restaurant discovery and reservations | Often paired with Wolfram or KAYAK in itinerary demos |
| Klarna | Klarna Bank AB | Shopping search and price comparison | Early example of fintech-AI integration |
| Speak | Speak Inc. | Language tutoring | Demonstrated educational scenarios |
| Slack | Salesforce | Workspace search and message summary | Early enterprise-oriented plugin |
| Shopify | Shopify Inc. | Product search across Shopify stores | Brought e-commerce inventory into ChatGPT |
| FiscalNote | FiscalNote Inc. | Real-time legal, political and regulatory data | Targeted enterprise compliance teams |
| Milo | Milo | Family scheduling and parenting helper | Showed personal-life automation |
Many of these vendors later returned to the AI ecosystem as Custom GPTs, MCP servers, or direct API consumers, but the original plugin versions stopped functioning after April 9, 2024.
On November 6, 2023, at its first DevDay conference, OpenAI announced GPTs. These are user-configurable versions of ChatGPT, each with a name, an avatar, a system prompt, optional knowledge files, and optional Actions that connect to external services. The Actions feature reused the OpenAPI-spec mechanism from plugins, but moved the integration inside a packaged assistant rather than exposing a free-form plugin list.
The GPT Store launched on January 10, 2024 and gave Custom GPTs the discoverability that the older plugin store had lacked. By 2025 the store contained millions of public Custom GPTs, including specialized assistants for coding, writing, design, education and shopping. From a developer's perspective, building a Custom GPT with Actions is similar to writing a ChatGPT Plugin: the same OpenAPI patterns apply. From a user's perspective, however, the experience is very different, because each GPT has a curated identity rather than being a generic plugin slot.
Key differences between the old Plugins system and the Custom GPT system are summarized below.
| Feature | ChatGPT Plugins (2023 to 2024) | Custom GPTs (2023 to present) |
|---|---|---|
| Status | Deprecated April 9, 2024 | Active, including the GPT Store |
| Discovery | Plugin store inside ChatGPT | GPT Store on chatgpt.com/gpts |
| Branding | Plugins shared the generic ChatGPT UI | Each GPT has its own name, icon, instructions |
| Tool definition | Manifest plus OpenAPI spec | System prompt plus Actions (OpenAPI) |
| Multiple tools | Up to 3 plugins enabled per chat | A single GPT may bundle multiple Actions |
| Authentication | None, API key, or OAuth | Same options |
| Knowledge files | Not supported | Supported (RAG over uploaded files) |
| Monetization | None | Originally previewed via revenue share for top GPTs |
For most practical purposes Custom GPTs occupy the niche that plugins once held. A developer who would have built a ChatGPT Plugin in 2023 would, in 2025, build either a Custom GPT, an MCP server, or a direct integration through OpenAI's Assistants API.
Most AI plugin systems share three architectural ingredients regardless of vendor:
The schema is presented to the model through a dedicated tools parameter or via the system prompt. The model emits a structured tool call when it decides one is needed; the host application intercepts it, executes it, and returns the result.
search_flights function.search_flights(destination="Tokyo", date="2026-06-01").This pattern is functionally identical across ChatGPT Plugins, Custom GPT Actions, Anthropic tool use, Gemini function calling, and MCP. The differences are in transport, packaging, distribution and discovery.
Two large families of AI plugins live outside the AI vendor's own product: browser extensions and editor extensions.
Browser plugins inject AI functionality into web pages or add a sidebar that calls an AI service. They are distributed through the Chrome Web Store, Firefox Add-ons, and equivalent stores for Edge and Safari. Notable examples include:
| Extension | Function | Notes |
|---|---|---|
| ChatGPT Sidebar | Adds a ChatGPT side panel to any website | One of the most installed AI browser plugins of 2023 |
| Glasp | AI-powered web highlighter and note taker | Uses an LLM to summarize highlights |
| Merlin | Cross-LLM sidebar (GPT, Claude, Gemini) | Lets users select a model per query |
| Monica | AI assistant with translation, summarization, and chat | Available on Chrome and Edge |
| WebChatGPT | Augments ChatGPT with web search results | Bridged the gap before native browsing returned |
| Compose AI | Inline writing assistant for email and documents | Productivity-focused extension |
As major browsers added native AI features, especially Microsoft embedding Copilot into Edge and Google integrating Gemini into Chrome, the third-party browser plugin ecosystem became less central, though many remain widely used.
The other large non-vendor plugin category lives inside code editors. AI assistants for programmers entered the mainstream with GitHub Copilot, released as a Visual Studio Code extension on June 29, 2021 and made generally available on June 21, 2022. Copilot's success demonstrated that AI features inside an editor were a fundamentally different product from a chat interface, and triggered a wave of follow-up extensions and dedicated AI editors.
The table below lists notable editor-side plugins and AI-native editors.
| Tool | Type | Host | Notes |
|---|---|---|---|
| GitHub Copilot | Plugin | VS Code, JetBrains, Vim, Visual Studio | First mainstream AI coding assistant |
| Cursor | AI-native fork of VS Code | Standalone editor | Supports its own extensions plus most VS Code extensions |
| Windsurf | AI-native editor by Codeium | Standalone | Includes its own model routing and tool layer |
| Tabnine | Plugin | VS Code, JetBrains, others | Older code completion product retrofitted with LLMs |
| Codeium | Plugin | VS Code, JetBrains, Vim, others | Free tier and enterprise variants |
| JetBrains AI Assistant | Plugin | JetBrains IDEs | First-party assistant from JetBrains |
| Continue | Open-source plugin | VS Code, JetBrains | Brings local or cloud models into the editor |
| Aider | Terminal CLI plus editor integration | Standalone with optional editor hooks | Pair-programming workflow against Git |
| Claude Code | Anthropic CLI | Terminal, with editor integrations | Coding agent with local file access |
Most modern editor plugins now build on the same tool-use machinery as MCP and function calling. Cursor, for example, lets users add MCP servers as first-class extensions, so a single MCP integration can serve both the editor and any other MCP-compatible client.
From 2024 onward the most important development in plugin architecture was the rise of the Model Context Protocol (MCP). Anthropic introduced MCP as an open standard on November 25, 2024 with reference implementations. Within months, MCP servers had been published for GitHub, Slack, Google Drive, Postgres, Brave Search, filesystem access, and dozens of other services.
MCP differs from earlier plugin systems in three important ways:
| Aspect | ChatGPT Plugins | MCP |
|---|---|---|
| Vendor scope | OpenAI only | Any client (Claude, Cursor, Windsurf, custom apps) |
| Hosting | Plugin must run on a public HTTPS endpoint | Local stdio, HTTP/SSE, or WebSockets |
| Standardization | OpenAI-defined manifest format | Open spec governed under an open license |
| Reuse | Each plugin is per-AI-product | One server can be reused across all MCP clients |
| Trust model | Service provider hosts the bridge | User can self-host the bridge for sensitive data |
By 2025, MCP servers had become the de facto way to extend Claude, Claude Desktop, Cursor, Cline, and many open-source clients. OpenAI itself added MCP support to several of its products in 2025. As of mid-2026 the public MCP registry lists hundreds of community-maintained servers, and large platform vendors such as Google, Microsoft, and Cloudflare ship official MCP integrations.
Anthropic also offers Claude Skills, a packaging format for reusable instructions and assets that load on demand. Skills sit alongside MCP rather than replacing it: MCP exposes external tools, while Skills bundle prompts, scripts, and resources that travel with Claude.
A fourth way to think about AI plugins is at the API level. OpenAI, Anthropic, Google, and xAI all offer first-class tool-use parameters, letting developers attach functions to any model call without going through the vendor's product UI.
On June 13, 2023, OpenAI added a functions parameter to the Chat Completions API. It was later renamed to tools and expanded to include parallel calls. Anthropic launched its own tools parameter for the Claude API in May 2024 with a JSON-Schema-based contract, and Google and xAI followed for Gemini and Grok.
The table below compares the three most widely used tool-use APIs as of 2026.
| Vendor | Parameter | First released | Schema format | Parallel calls |
|---|---|---|---|---|
| OpenAI | tools (formerly functions) | June 13, 2023 | JSON Schema | Yes |
| Anthropic | tools | May 30, 2024 | JSON Schema | Yes |
tools (Gemini) | December 2023 | OpenAPI 3.0 subset | Yes |
For many production systems, raw API tool use has effectively replaced packaged plugin marketplaces. A backend service can attach exactly the tools it needs to a model call, audit every invocation, and run within the developer's own infrastructure rather than depending on the AI vendor's plugin sandbox. This pattern is the dominant approach behind enterprise AI assistants, autonomous agents like Devin, and most modern AI applications shipped in 2025 and 2026.
The table below summarizes how the major AI plugin formats compare in terms of openness, host coupling, and current status.
| Format | Year | Openness | Host coupling | Status (May 2026) |
|---|---|---|---|---|
| ChatGPT Plugins | 2023 | Proprietary | OpenAI only | Deprecated |
| OpenAI tools / function calling | 2023 | API standard | OpenAI only | Active |
| Custom GPTs (Actions) | 2023 | Proprietary | OpenAI only | Active |
| Anthropic tool use | 2024 | API standard | Anthropic only | Active |
| Gemini function calling | 2023 | API standard | Google only | Active |
| Model Context Protocol | 2024 | Open standard | Any compliant client | Active |
| Editor extensions | 2015 onward | Open | Editor only | Active |
| Browser extensions | 2003 onward | Open | Browser only | Active |
| Claude Skills | 2025 | Anthropic | Claude only | Active |
In practice, building a serious integration in 2026 typically means choosing some combination of MCP, native tool use, and an editor or browser extension. Pure ChatGPT-Plugin-style integrations are no longer a meaningful option.
The rapid rise and retirement of ChatGPT Plugins between 2023 and 2024 left several durable lessons for the AI industry.
See also: List of ChatGPT Plugins
The table below preserves the original entries from the previous version of this article. All listed plugins ceased operation on April 9, 2024 along with the rest of the ChatGPT Plugins system.
| Plugin | Image | Model | Category | Release Date | Description | Available | Working |
|---|---|---|---|---|---|---|---|
| BuyWisely (ChatGPT Plugin) |  | GPT-4 | E-Commerce | May 20, 2023 | Compare Prices & Discover the Latest Offers from thousands of online shops in Australia. | No (deprecated April 9, 2024) | No |
| Instacart (ChatGPT Plugin) |  | GPT-4 | Food Delivery | May 20, 2023 | Ask about recipes and discover the ingredients you need, then get it all delivered from local stores. | No (deprecated April 9, 2024) | No |
| Speechki (ChatGPT Plugin) |  | GPT-4 | Technology | May 20, 2023 | The easiest way to convert texts to ready-to-use audio - download link, audio player page, or embed! | No (deprecated April 9, 2024) | No |