Interlinked.

Comparisons · How agents get their tools

ChatGPT started the plugin era. Even OpenAI moved to MCP.

For a year and a half, giving an AI a new skill meant building it the way ChatGPT did in 2023: a plugin, wired to one vendor's cloud, usable only inside that one app. Then Anthropic published the Model Context Protocol (an open standard any agent can speak) and the math inverted. Build a capability once, and every MCP agent can use it: Claude Code, Cursor, and by 2025, ChatGPT itself. Here is the difference between the two models, drawn out, and why a single install can hand the same file search to 19 different AI clients.

2023 · CHATGPT PLUGIN
one app, one vendor: rebuilt for the next
2024 · MCP SERVER
every agent at once: one server, shared

The short version

  • 01ChatGPT plugins (2023) let one app (ChatGPT) call outside tools. Every other assistant needed its own separate, incompatible format.
  • 02MCP flips it: one server exposes a capability, and every MCP agent (Claude Code, Cursor, ChatGPT and more) can use it. Build once, reach all.
  • 03By 2025, even OpenAI, which started the plugin era, had added MCP support to its own products. The neutral standard won.
  • 04Plugins tie a tool to a vendor; MCP ties it to a protocol. Switch editors and an MCP tool comes with you; a plugin gets left behind.
  • 05Interlinked Files is one MCP server: a single install reaches 19 AI clients and answers file search in microseconds, 85 µs across 4.47M files.

Definitions first

What each one actually is.

CHATGPT PLUGINS

ChatGPT plugins were OpenAI's 2023 system for giving ChatGPT access to outside tools: third-party services the model could call mid-conversation, defined to OpenAI's own spec and usable only inside ChatGPT.

MODEL CONTEXT PROTOCOL

MCP is an open standard that lets any AI agent connect to any tool or data source through one shared interface: write the connection once, and every agent that speaks MCP can use it, no per-vendor rebuild.

HOW WE GOT HERE: THE TWO-YEAR SWITCH
  1. Mar 2023
    OpenAI launches ChatGPT plugins in beta. Tools live inside ChatGPT, and nowhere else.
  2. Late 2023 to 2024
    OpenAI shifts to GPTs and the GPT Store, and winds the original plugins down.
  3. Nov 2024
    Anthropic introduces the Model Context Protocol as an open standard, and open-sources it.
  4. 2025
    Claude Code, Cursor and a wave of agents ship native MCP; OpenAI adds MCP to its own products.
  5. Today
    One MCP server reaches 19 AI clients in a single install. New agents that speak MCP just work.

The clearest signal is the last step: the company that popularized AI plugins now speaks MCP too. When the incumbent adopts the neutral standard, the debate is over, not because plugins were bad, but because a tool you build for whatever agent you happen to use beats a tool locked to one. For the full primer, see what is an MCP server.

The two models, drawn

Three vendors, three formats. Or one server, every agent.

On the left, the plugin model: each vendor invents its own format, and each capability is trapped inside the one app that hosts it. Want “find files” in a second assistant? Build it again, its way. On the right, MCP moves the capability outside every app into one small server; each agent (including ChatGPT, which adopted MCP in 2025) connects to the same one.

THE PLUGIN MODELone vendor, one format, locked inChatGPTFIND FILESGPT pluginruns only hereCopilotFIND FILESCopilot ext.runs only hereVS CodeFIND FILESVS Code ext.runs only here3 FORMATS · NONE PORTABLETHE MCP MODELone server, shared by every agentClaude CodeChatGPTCursor+16 moreINTERLINKED FILESONE MCP SERVERbuilt once · installed everywhere in one step

The left side is why tool makers used to pick a favorite app and ignore the rest: every new host meant a fresh integration to write and maintain. The right side is why that math changed: the server is written against the protocol, not against any one app, so the agent on the other end is interchangeable. This is the same shift covered in the backbone of the LLM economy.

Side by side

Six questions, two very different answers.

The questionChatGPT-style pluginMCP server
Which apps can use itOnly its host, e.g. ChatGPT itselfEvery MCP agent on the machine
Who defines the formatEach vendor, privatelyOne open standard, shared
To support a new agentBuild a whole new pluginIt already works
You switch agentsLeft behindComes with you
Where it runsThe vendor's cloud, behind the appIts own process, on your machine
Do your files leave the machineOften: the tool is vendor-hostedNo: local server, offline by default

None of this makes plugins useless: a deep, app-specific extension is still the right tool when you want to reshape one editor. But when the job is handing a capability to whatever agent you happen to be using, the plugin model asks you to bet on which vendor wins. MCP lets you skip the bet, and keeps your files local while it does.

Measured, not claimed

What one shared server buys.

6m 57s → 16ms
Claude Code finding one file without an index, then with it: the length of a song, down to a blink. 71 tool calls become 1.
412,000×
one filename lookup vs VS Code's Ctrl+P, same drive: 85 µs against 35 seconds of spinning.
~58% → ~0%
share of an agent's session tokens spent hunting for files: roughly half the bill, reclaimed.
The searcherWhat ranTimevs Interlinked
Interlinked Filesone indexed filename lookup85 µsbaseline
VS Code (Ctrl+P)same file, same drive35 s412,000×
Windows Searchmedian of 20 queries67 s~480,000×
ripgrepsame content query, same corpus93.8 s58,625×
Claude Code (no index)one file, agentic archaeology6 m 57 s~7,200,000×

Methodology: Ryzen 9 9950X3D · 64 GB DDR5 · NVMe · Windows 11 · 4,470,000 files. Each ratio is against one indexed Interlinked lookup on the same corpus: filename lookups are microsecond-scale (85 µs single, 139 µs median of 20 queries); the ripgrep row is content-to-content (93.8 s → 16 ms; 58,625× across the query set); the Claude Code row is a measured single-file agent run with no index, listing directories until it found the file. Every one of those numbers is identical no matter which of the 19 clients made the call.

The file-search MCP server

Interlinked Files is one server every agent shares.

It exposes four tools (filename search, content search, both at once, and a knowledge lookup that returns everything known about one file, folder, or repo) over the same protocol Claude Code, Cursor and ChatGPT already speak. There is no ChatGPT plugin to publish, no per-vendor format, no account for the search itself: one signed installer configures 19 AI clients in a single step, and the moment a new MCP agent ships, it works too. Backed by an always-fresh index of every file on your PC (every repo, every document, even the git-ignored files a vendor-hosted plugin never sees), measured on a real 4.47-million-file machine. It stays on your machine, offline, sitting around 44 MB at rest (less than a single browser tab), with the index taking under 1% of the drive. Save a file and it is searchable again in about a millisecond.

19
AI clients auto-configured by one install: Claude Code, Cursor, Copilot, Codex, Windsurf, Zed, Cline and more, plus ChatGPT's own MCP support
1 install
not one plugin per vendor: the moment a new MCP agent ships, it already works
85 µs
one filename lookup across 4.47M files (139 µs median of 20 queries, identical from every agent)

ONE SERVER TO INSTALL. NINETEEN AGENTS THAT GET IT. ZERO PER-VENDOR PLUGINS.

Common questions

MCP vs ChatGPT plugins, answered.

Are ChatGPT plugins still available?
OpenAI wound down the original ChatGPT plugins through 2024 and folded that functionality into GPTs. Today most tool access across the ecosystem flows through MCP: the open standard OpenAI itself adopted in 2025.
Is MCP an OpenAI thing or an Anthropic thing?
Neither owns it. MCP is an open standard introduced and open-sourced by Anthropic in late 2024. Claude Code, Cursor, and OpenAI's own products all speak it. That vendor-neutrality is the whole point.
Do I need a separate setup for each agent?
No. That is exactly what MCP removes. One server is installed once, and every MCP client on the machine can call it. Interlinked Files configures 19 clients in a single install.
Does an MCP server send my files to the cloud?
It doesn't have to. Interlinked Files runs entirely on your machine: the index and every query stay local and offline, with no account or key for the search itself. A ChatGPT-style plugin, by contrast, is hosted in the vendor's cloud.
What does Interlinked Files actually expose over MCP?
Four tools: filename search, content search, both at once (answering across all 4.47M files on the test machine in microseconds, identical no matter which agent asks), plus a knowledge lookup that returns everything known about one file, folder, or repo: its machine-wide connections and notes.

Ready to wire it up? The step-by-step is in how to set up an MCP server in Claude Code.

Do this today

Skip the plugins. Install the layer.

1

Install once. One signed installer auto-configures 19 AI clients: Claude Code, Cursor, Copilot, Codex, Windsurf, Zed, Cline and the rest. No plugin to publish, no per-vendor JSON to edit.

2

Let it index. Filename search works immediately; content search fills in behind it. At rest the whole engine sits around 44 MB (less than a browser tab), and stays entirely on your machine.

3

Ask from any agent. Whichever agent you happen to be in, the tools are already there and answer identically. Switch editors tomorrow and there is nothing to reinstall.

4

It stays fresh. Save a file and it's searchable in about a millisecond (under 30 in the worst case). The index rides your changes; you never re-index by hand.

Everything local is free forever, no card. Hosting starts at $5.99/mo. Download for Windows.

Measured July 2026 · Ryzen 9 9950X3D · 4.47M files · Windows 11← All research