Lesson 8.88.8 The same setup on today's tools

In everyday life. Two phone companies both let you keep a contact list. One syncs it automatically, updating it by itself. The other makes you import it again after each change. The list is the same.

What you will learn. How to set up the same project on both AI vendors, from their own help pages, and the two differences that change your setup. You practice it in Tasks 3 and 13 of the lab.

Why it matters. The ideas are the same on both AI vendors, but two settings are not. Get them wrong, and a copy goes out of date, or memory from your other chats gets into the project.

Anthropic makes Claude, and OpenAI makes ChatGPT. They are the two AI vendors, the companies that make the AI products in this book. On either one, you set up the same project, a workspace for one body of work. The steps are the same. The settings differ.

The comparison. Both AI vendors offer projects with instructions and files. Both can keep a project's memory apart from your other work: on ChatGPT, once you choose project-only memory. Both let you read and correct saved memories. Two differences change how you set up:

1. Keeping copies current. A project holds copies of your approved knowledge. A Google Doc in a private Claude project stays synced, refreshed from time to time when you open the project.¹ The Google Drive app inside a ChatGPT project fetches files when asked, and does not sync them in advance.² On either AI vendor, replace uploaded files by hand after an approved change. On ChatGPT, also check that answers from fetched files cite the new approval.
2. Memory scope, which memories a project uses. Every Claude project keeps separate memory.³ A ChatGPT project uses your saved memories, unless you choose project-only memory.² Choose project-only memory for an AI Worker's project.

Open source adds no third difference to this setup: it changes where memory and state are kept, not what they are. Each kind of information still goes where Section 8.3 says, and on OpenCode a standing rule goes into AGENTS.md.

Figure 8.9. The same setup on both AI vendors, as verified 7 October 2026. The gold rows are the differences that change your setup.

What stays the same. An answer can come from five places, on either AI vendor or an open-source tool:

• Context, what the worker sees in this conversation, is temporary.
• Memory is continuity, never authority. It helps the worker carry on from past chats, but it never has the final word.
• A matter's history, the story of one piece of work such as an invoice, is kept in SSoR, the State System of Record, not in a chat.
• Approved knowledge is written once, in the KSoR, the Knowledge System of Record, and a named person keeps every copy current.
• Current state, where things stand now, is read from its own system, such as the accounting system. In Part IV of this book, DSoR, the Data System of Record, does that reading.

For reference. The two AI vendors' boxes below cover the same seven settings, in the same order. The open-source box covers the settings its tools match, and how to run them safely.

Anthropic, as verified 7 October 2026.

• Context window. The context window is everything the model can use for its next reply, counted in tokens, pieces of words. On paid plans, the newest models have a 1M-token (one million token) window. Near the limit of the window, Claude summarizes earlier messages and keeps the full history to refer back to. This summarizing needs code execution, a setting that lets Claude run code, turned on.⁴ Long chats use more of your usage limit, the amount of use your plan allows.⁵
• Instructions. Account-wide instructions in Settings apply to every conversation.⁶ Project instructions apply to one project.⁷ On Team and Enterprise, organization owners can set organization instructions, which take priority where they conflict with a member's own instructions.⁸
• Project knowledge. Project files are available in every chat in the project.⁷ On paid plans, when a project's files grow near the window's limit, it switches to retrieval automatically. Claude then searches the files and reads only the parts it needs.⁹
• Synced sources. Google Docs added from Google Drive to chats and projects stay synced to the latest version. In projects, this works only in private projects, ones not shared with others,¹⁰ and the content refreshes from time to time when you open the project.¹
• Memory. Claude saves memory as topics you can read, edit and delete in Settings. Each project has its own memory. Memory can be turned off for one chat, inside or outside a project, before the first message.³ A second setting, Search and reference chats, lets Claude search past chats on paid plans.³ In a run on 3 November 2026, a chat with memory off still found an older chat while that setting was on. Incognito chats, outside projects, are not saved to memory.¹¹ On Team and Enterprise, owners decide whether memory is available.³
• Connectors. Connectors are links from Claude to other apps and services. They reach them with your own permissions. If you cannot open a file, the connector cannot either, unless it was set up with one shared login that many people use.¹²
• Sharing. On Team and Enterprise, a project can stay private. It can also be shared with chosen people, with Can view or Can edit access, or be opened to the whole organization.¹³

OpenAI, as verified 7 October 2026.

• Context window. OpenAI lists context windows by plan and mode on its pricing page.¹⁴ OpenAI's help pages do not say how ChatGPT handles older messages when a chat grows long.
• Instructions. Custom instructions, ChatGPT's account-wide instructions, apply to every chat in your account.¹⁵ Project instructions apply only within the project and override, or take priority over, your custom instructions.²
• Project knowledge. A project holds chats, files and instructions. The file limit is 5 per project on the Free plan, 25 on the Go and Plus plans, and 40 on the Pro, Business, Enterprise and Edu (education) plans.²
• Synced sources. In a private project, you can add a Google Drive file or folder as a source by pasting its link. The Google Drive app, a connector, does not sync when added inside a project. It can still search and open files when asked.² Sync indexes content for search. It is now set up only by workspace admins, for supported providers.¹⁶
• Memory. Saved memories are details you ask ChatGPT to remember. Where that behavior is available, they also include details ChatGPT saves as useful. A second setting, Reference chat history, uses past chats. You can review, correct and delete saved memories.¹⁷ A project's own memory shows no list. To stop project memory using a chat, delete the chat or, where settings allow, move the chat to another project.² Temporary chats do not create or update memories.¹⁸ A project can use project-only memory, which ignores saved memories and chats outside the project. Shared projects always use project-only memory.² As verified 23 October 2026, a dot, OpenAI's always-on agent, shares memory with ChatGPT both ways. For now, you cannot view, delete or change a dot's individual memories. Deleting the dot deletes its context.¹⁹
• Connectors. In ChatGPT, connectors are now called apps.²⁰ On Business, Enterprise and Edu, the Company Knowledge plugin, an add-on, searches connected apps within each user's existing access. When an answer includes citations, you can open them and check.²¹
• Sharing. A shared project gives members edit access or chat access.²

Open source, as verified 23 October 2026.

• Instructions. OpenCode reads standing instructions from files named AGENTS.md. It loads one global file for all your work, then each AGENTS.md in your working folder and the folders above it.²²
• Memory. OpenClaw remembers things by writing plain Markdown files in the agent's workspace, a folder on the machine that runs it.²³ Its session transcripts, the records of what was said, are kept in a database file on the same machine.²⁴ Omnara keeps each agent as an event log, a running list of what it did.²⁵ Every step is saved, and that record is the agent's state. So the agent can be picked back up after a crash or restart. If you host Omnara yourself, the record sits in your own database.²⁶ Kept on your own hardware, none of this is the matter's case file.
• Connectors. OpenCode connects to MCP servers, on your own computer or remote. MCP (Model Context Protocol) is a standard way to connect AI apps to other systems.²⁷ KSoR's served route, which Part IV builds, is an MCP server too.
• Running it safely. Memory kept on your machine still goes to the model's provider when it enters the context, unless the model runs on your own computer.²⁸ So keep anything sensitive out of memory files, as you would with a vendor's memory.

Skills, as verified 12 October 2026. Skills sit outside any one project, so the boxes above leave them out. Section 8.5 explains them.

• Anthropic. Skills work on every Claude plan, with code execution turned on. You turn them on under Customize > Skills.²⁹ On Team and Enterprise, an owner can give a skill to everyone.³⁰
• OpenAI. Skills are under Plugins, on the Skills tab, on the Business, Enterprise, Healthcare and Edu plans.³¹ On other plans, keep the procedure in the project instructions.²

Plugins, as verified 17 October 2026. A plugin is a package of several kinds of setup, such as skills and connectors. Like a skill, it sits outside any one project. Section 8.9 says to read every part inside a plugin before you turn it on. Both AI vendors show you those parts before you add one.

• Anthropic. A plugin can hold skills, connectors, commands and agents. You add one at Customize > Plugins, on the Discover tab. Select a plugin to read what it contains, then select Add. Anthropic reviews the plugins listed in its directory, but not a plugin you add from another source or upload yourself. Adding a plugin does not connect anything. Its Connectors tab shows each connector it uses, and whether that connector is connected on your account.³² Plugins are for the paid plans: Pro, Max, Ultra, Team and Enterprise, as of 31 October 2026.³³ Anthropic also lists connectors and plugins on a website, the Claude Marketplace.³⁴ On Team and Enterprise, an owner chooses how members get each plugin the organization offers. It can be available to install, installed by default, or required, so members cannot turn it off. It can also be hidden.³⁵
• OpenAI. Open Plugins in the ChatGPT sidebar to reach the plugin directory, which every plan has. Whether you can install a given plugin depends on your plan, workspace, role and region. A plugin can hold skills and apps, ChatGPT's name for connectors. Select a plugin and review its description, skills, apps and setup requirements before you select Install plugin. Installing can start each app's sign-in, and it cannot get around the provider's or the workspace's permissions. In a workspace, admins manage plugins in the Admin console. The plugin's settings decide whether members can have it, and each app's settings decide what that app may reach. On Enterprise and Edu, admins can also set plugin access by role.³⁶