Chapter 4 on today's products
- Approved by
- Panaversity ·
This page backs 4.8. The lesson teaches the two differences that change Brightline's design: how the worker reaches the KSoR, and where the trigger is set. This page holds the products' own details, layer by layer, with their sources and the date each was checked. It also holds today's products for the choices in 4.10.
Products change every few weeks. Check the date on each box before you rely on a line, and open its source if the date is old.
Each product, layer by layer
The three boxes follow Figure 4.5 in 4.8 from the top down: Anthropic, OpenAI and open source. The products in the boxes will change. The two differences the lesson teaches will not.
Anthropic, as verified 18 November 2026.
- Channels: the Claude apps on web, desktop and mobile,1 Claude Tag in Slack, in beta on Team and Enterprise plans, where Claude acts under its own identity,2 and the API.3
- Runtimes: tasks in a Claude conversation, which Anthropic calls Cowork, for everyday delegated work. Each cloud session gets its own sandbox, a small virtual machine that is destroyed when the session ends. After about five idle minutes, it is saved as an encrypted snapshot, and a new message restores it into a fresh machine. Snapshots expire 30 days after the session was created. Session history lives with the account.4 Claude Code, the coding harness, on your own computer or as a cloud session, on Pro, Max, Ultra and Team plans and on Enterprise with the right seats, where each session runs in its own machine. An idle machine is reclaimed, and reopening the session gives a fresh one with only the conversation restored.5 The Claude Agent SDK (software development kit), which you host yourself. Claude Managed Agents, in beta, is a pre-built agent harness that you can configure. It runs in managed infrastructure, in an Anthropic-managed cloud sandbox or in one on your own infrastructure. Its sessions are stateful by design: conversation history, sandbox state and outputs are stored on the server between interactions.3
- Memory: Claude generates memory from your chats, and you view and edit it in Settings. It is on by default on the Free, Pro, Max and Ultra plans. On Team and Enterprise plans, owners control whether memory is available. In most organizations created on or after October 31, 2026, it starts on for each member. In older ones it is off until they turn it on. Each project has its own memory, and also uses your account memory unless you turn that off (8.8). Incognito chats are not saved to memory.6
- Reaching a KSoR: custom connectors over remote MCP, in Claude, its tasks and Claude Desktop, on every plan. The Free plan allows one.7
- Triggers: scheduled tasks run hourly, daily, on weekdays, weekly or on demand, on the paid plans. Each scheduled task runs as its own session. They are time-based.8
- Standing runtime: Claude Duet, Anthropic's standing worker. It is a research preview with a waitlist, starting with Pro, Max and Ultra subscribers in the US. For now it is its own app, separate from the main Claude app. It has a virtual machine with its own browser, so your laptop and phone can be off. It checks your connected accounts every few hours. Before it sends a message, makes a payment, signs a document or deletes anything, it waits for your approval. How long its machine and the files on it last is not stated.9
- Plans: Pro is $20 a month, Max is $100 and Ultra is $200.10 The Free plan has chat, up to five projects and memory.1 It allows one custom connector.7 Handing off and scheduling tasks, and Claude Code, start on Pro.1
- Export: a data export holds conversation data and the user data for the account, and any voice recordings you shared with Anthropic. It does not hold the files you uploaded. Nothing names a project's instructions.11
OpenAI, as verified 21 November 2026.
- Channels: the ChatGPT apps, @ChatGPT in Slack and Teams, and the API.12
- Runtimes: tasks in a ChatGPT conversation, which OpenAI calls Work, for everyday delegated work. Codex, the coding harness. The OpenAI Agents SDK, an open-source framework you host yourself.13 The Agents API, in beta, is an OpenAI-managed Codex agent. OpenAI manages the session and the coordination between model and tools, with an OpenAI-hosted sandbox as one option.14
- Memory: saved memories, which you ask ChatGPT to keep or which ChatGPT saves itself, and reference chat history, both controlled in Settings.15 A project can use project-only memory, and a shared project always does.16 Temporary Chat does not create or update memories, and set to Unpersonalized it reads none.15
- Reaching a KSoR: custom MCP servers, added from Plugins. OpenAI documents full MCP support for Business and Enterprise or Edu workspaces on the web, where admins and owners create and publish the connections. It documents read and fetch access for Pro users. Its page does not list the Free plan.17
- Triggers: scheduled tasks run once or on a schedule, on every plan. Event-triggered tasks in ChatGPT respond to Gmail, Slack or GitHub activity, on Plus and above.18
- Standing runtime, verified 21 November 2026: a dot, OpenAI's always-on agent with its own cloud computer, for ongoing work. How long that computer and the files on it last is not stated.19
- Export, verified 21 November 2026: the export ZIP holds chat history and other account data, and is not available in Business or Enterprise workspaces.20 Deleting a project removes its chats, its instructions and the files stored only in it.16
Open source, as verified 22 October 2026.
- Channels: OpenClaw meets you in the chat apps you already use, such as Discord, iMessage, Slack, Teams, Telegram and WhatsApp.21 OpenCode has a terminal interface, a desktop app and a web app.22 Omnara's agents work in Slack and Discord, where they reply, ask questions and ask for approvals.23
- Runtimes: OpenCode, an open-source coding agent, the counterpart of Claude Code and Codex.22 OpenClaw, an open-source assistant that runs on your own computer, for ongoing work. Its tools run on that computer itself unless you set up a sandbox.21 Omnara, which its makers describe as an alternative to Claude Managed Agents and OpenAI's Agents API that you can host yourself. You run it with Docker, or use Omnara Cloud. Its agents run in sandboxes from outside providers, or on your own laptop, server or container. There, a small program connects out to Omnara, so nothing needs to connect in.23
- Models: you choose the model. OpenCode works with many providers, and with models that run on your own computer, such as through Ollama.24
- Memory: the worker's state stays on hardware you control. OpenClaw runs on your own computer.21 Omnara saves every step of every agent, and a copy you host yourself keeps that history in your own database.23 A saved history is still not a case file.
- Reaching a KSoR: OpenCode connects to MCP servers, the same standard the two AI vendors use.22
- Triggers: Omnara's timed triggers can message a running agent or start a new one.23
- Plans: OpenClaw is free under the MIT license,21 and Omnara is free to host yourself under the Apache 2.0 license.23 You pay for the model, unless you run one yourself, and for the computers. OpenCode Go is an optional $10-a-month plan for coding models.22
- Running it safely: OpenClaw treats incoming messages as untrusted, and its chat channels make unknown senders pair with it first. Its README points to its security guide before you connect other people or reach it from outside your network.21 Omnara's local settings are insecure on purpose, so a real deployment follows its self-hosting guide.23
Products for the choices in 4.10
Products for 4.10, as verified 22 October 2026, with the data parts checked again on 23 October, Anthropic's lines on 18 November, and OpenAI's on 21 November. 4.10 asks what changes when a team or another company starts to use a worker. These are today's products for each of its choices. The products will change. The choices will not.
| The choice | Anthropic | OpenAI | Open source |
|---|---|---|---|
| A team uses the worker | Claude Tag in Slack, under its own identity (beta) | Workspace agents, in ChatGPT Business, Enterprise and Edu25 | Omnara's agents, in Slack or Discord |
| A standing worker | Claude Duet (research preview) | A dot | OpenClaw, on your own hardware |
| A worker a vendor sets up for a company | None named | Specialist dots, in a preview with focused enterprise pilots26 | None in this book |
| A service the vendor runs | Claude Managed Agents (beta) | The Agents API, in public beta since 10 September 202627 | Omnara Cloud |
| A toolkit you run | The Claude Agent SDK, in a process you operate28 | The OpenAI Agents SDK, in Python or TypeScript29 | Omnara, on your own servers |
| One call to a model | The Messages API | The Responses API | A provider's API, or a model on your own computer |
-
Where the data goes, and traces. These are under Data and traces, below.
-
Where a graph's rules live. Agent Builder, OpenAI's tool for drawing agent graphs, is being retired, and it shuts down on 30 November 2026. OpenAI points its users to the Agents SDK or to workspace agents.30 A graph drawn only inside that tool goes with it. Claude Managed Agents offers subagents and workflows that run many agents, in beta.31
-
MCP's new version. MCP's version of 28 July 2026 changes how a connection keeps state.
- A server that needs state between calls passes a handle as an ordinary tool input.
- A server can ask for more input by returning a request for it.
- Long-running tasks moved out of the core, into an official extension.32
- Anthropic says support is rolling out across Claude products, and its post names the Claude apps and the Claude Platform.33
So a connection that carries across a move still needs retesting.
Data and traces
These details back the third item, a written answer to where the data goes, in 4.10.
- Where the data goes. Zero Data Retention (ZDR) means the vendor keeps no copy of your data after it answers.
- Claude Managed Agents is not eligible for ZDR, unless you host it yourself. It is not eligible for HIPAA Business Associate Agreement coverage, which US health data needs.3
- With a sandbox on your own infrastructure, its files and processes stay there. But the conversation history, tool results, skills and memory stores still reach Anthropic.34
- For the Messages API, ZDR is turned on for each organization, through Anthropic's sales team. Some models are excluded.35
- Where Anthropic's API, Amazon Bedrock or Google Cloud can keep the model's work is under Keeping data in one region, in Chapter 10 on today's products.
- The Agents API keeps data only in the United States, and it does not support ZDR, even with a sandbox on your own infrastructure.36 Its records of each agent are kept until you delete them.37
- The Responses API is eligible for ZDR, with limits.37 It can keep data in a region you choose. OpenAI's data residency guide lists the regions, among them the United States, Europe and the United Arab Emirates.38
- With open-source tools, the runtime and its records stay where you run them. Your prompts still go to the model's provider, unless the model runs on your own computer.24
- Traces. A trace is a log of what an agent did, step by step. Like a run's log, it shows what was attempted, not what really happened (4.10). It can also have gaps, so it does not replace the case file.
- Anthropic's export of traces for the Claude API is in beta. It sends them to a monitoring tool you choose. By default it sends only metadata, such as identifiers, timing, status and token counts. Prompts and responses go only if you choose full content. Anthropic's page warns that the tool then stores that content under its own settings for access and retention. Under a HIPAA, FERPA or K-12 restriction, the API sends only metadata. Anthropic's page also says exported traces are not a complete record: delivery is best effort, so a step can be lost.39
- The OpenAI Agents SDK traces by default, to OpenAI's own servers. By default its traces include the model's inputs and outputs, and a setting leaves them out. It can also send traces to other tools. Tracing is not available under ZDR.40
Check yourself
Question 1 / 3
0 answered
Brightline wants Anthropic to host a worker that Brightline configures. Which product fits, as of 18 November 2026?
Sources
- Plans & pricing, Anthropic.↑↑2↑3
- What is Claude Tag?, Claude Help Center.↑
- Claude Managed Agents overview, Claude Platform Docs.↑↑2↑3
- Unified Claude Security Best Practices, version 1.3.6, Anthropic Trust Center, 7 November 2026.↑
- Use Claude Code in the cloud, Claude Code docs.↑
- Use Claude's chat search and memory to build on previous context, Claude Help Center.↑
- Get started with custom connectors using remote MCP, Claude Help Center.↑↑2
- Schedule recurring tasks in Claude Cowork, Claude Help Center.↑
- Introducing Claude Duet, Anthropic, 21 October 2026.↑
- Plans & pricing, Anthropic, accessed 18 November 2026.↑
- Export your Claude data, Claude Help Center.↑
- Using @ChatGPT in Slack and Microsoft Teams, OpenAI Help Center.↑
- OpenAI Agents SDK, OpenAI.↑
- Agents API (beta) FAQ, OpenAI Help Center.↑
- Memory in ChatGPT, OpenAI Help Center.↑↑2
- Projects in ChatGPT, OpenAI Help Center.↑↑2
- Connecting custom MCP servers in ChatGPT, OpenAI Help Center.↑
- Scheduled tasks in ChatGPT, OpenAI Help Center.↑
- Introducing dots, OpenAI, 29 September 2026.↑
- Exporting your ChatGPT history and data, OpenAI Help Center.↑
- OpenClaw, OpenClaw Foundation, README on GitHub.↑↑2↑3↑4↑5
- OpenCode v2 documentation, OpenCode.↑↑2↑3↑4
- Omnara, README on GitHub.↑↑2↑3↑4↑5↑6
- Providers, OpenCode v2 documentation.↑↑2
- Introducing workspace agents in ChatGPT, OpenAI.↑
- Introducing dots, OpenAI, 29 September 2026.↑
- Changelog, OpenAI developer docs.↑
- Agent SDK overview, Claude Code docs.↑
- Agents SDK, OpenAI developer docs.↑
- Deprecations, OpenAI developer docs.↑
- Multiagent orchestration, Claude Platform Docs.↑
- Key changes, MCP specification 2026-07-28, Model Context Protocol.↑
- Bringing MCP 2026-07-28 to Claude, Anthropic, 28 July 2026.↑
- Choosing an agent environment, Claude Platform Docs.↑
- API and data retention, Claude Platform Docs.↑
- Agents API, OpenAI developer docs.↑
- Data controls in the OpenAI platform, OpenAI developer docs.↑↑2
- Data residency, OpenAI developer docs.↑
- OpenTelemetry tracing for the Claude API, Claude Platform Docs.↑
- Tracing, OpenAI Agents SDK.↑
- Owner
- Panaversity