claude-docs-20261022T180711Z / page 11 of 11

claude-science/example-gallery/soc-schelling-small Changed · +1 / -1 lines

from line 8

88

99## Prompt

1010

11Copy the prompt and paste it into a new session in [Claude Science](/docs/claude-science/overview). By default, Claude asks with a [permission card](/docs/claude-science/core-concepts#permission-cards) before it runs code or installs packages.

11Copy the prompt and paste it into a new session in [Claude Science](/docs/claude-science/overview.md). By default, Claude asks with a [permission card](/docs/claude-science/core-concepts.md#permission-cards) before it runs code or installs packages.

1212

1313```text Prompt wrap theme={null}

1414In 1971 the economist Thomas Schelling showed, with chips on a checkerboard, that a city can segregate sharply even when nobody asks for more than an even mix. Recreate the textbook computer version, which segregates at even milder demands. Fill a 50 × 50 grid at random with 1,125 blue and 1,125 orange households, leaving 250 cells empty. A household is content if at least 30 % of its occupied neighboring cells (the 8 around it, fewer at the edges) hold its own color. With 8 occupied neighbors, that means at least 3 of its own color, so it accepts 5 of the other. Each discontent household moves to a random empty cell; repeat until everyone is content. Show the grid before and after and measure the average share of like-colored neighbors: it starts near 50 %. Where does it end? How many points is that above the random board, and how many above the 30 % they asked for?

claude-science/mac-app-window Changed · +2 / -2 lines

from line 6

66

77> [!NOTE]

88>

9> To manage Claude Science on your organization's Macs, see [Manage Claude Science on devices](/docs/claude-science/manage-on-devices#macs-where-claude-science-opens-in-its-own-window).

9> To manage Claude Science on your organization's Macs, see [Manage Claude Science on devices](/docs/claude-science/manage-on-devices.md#macs-where-claude-science-opens-in-its-own-window).

1010

1111## What changes when Claude Science switches to its own window

1212

from line 23

2323

2424## macOS permissions for Claude Science

2525

26macOS treats Claude Science in its own window as a different app from the browser version, so macOS asks again for some permissions, such as access to your Documents or Desktop folder. Dictation needs microphone access. If dictation doesn't work, see [Dictation](/docs/claude-science/troubleshooting#dictation) in Troubleshoot Claude Science.

26macOS treats Claude Science in its own window as a different app from the browser version, so macOS asks again for some permissions, such as access to your Documents or Desktop folder. Dictation needs microphone access. If dictation doesn't work, see [Dictation](/docs/claude-science/troubleshooting.md#dictation) in Troubleshoot Claude Science.

2727

2828Access you turned on yourself, such as Full Disk Access, doesn't carry over from the browser version. To turn it on again, go to **System Settings > Privacy & Security**, choose the kind of access, and turn on the **Claude Science** toggle. If **Claude Science** isn't listed, click the add button (+) and choose **Claude Science** in the **Applications** folder. Do the same if a **Claude Science** entry is already on but Claude Science still can't reach your files, because that entry can be the browser version's.

2929

claude-science/monitor-usage Changed · +1 / -1 lines

from line 2

22

33> Claude Science uses the same seat as the rest of claude.ai. Track its adoption and usage in Analytics and through the Admin API.

44

5Claude Science uses the same seat as the rest of claude.ai, and its usage counts the same way as Claude Code and Cowork usage (see [Plans and usage](/docs/claude-science/overview#plans-and-usage)). You can track adoption in Analytics and through the Admin API.

5Claude Science uses the same seat as the rest of claude.ai, and its usage counts the same way as Claude Code and Cowork usage (see [Plans and usage](/docs/claude-science/overview.md#plans-and-usage)). You can track adoption in Analytics and through the Admin API.

66

77## Analytics

88

claude-science/remote-compute-clusters Changed · +1 / -1 lines

from line 4

44

55Connect a machine you can reach over SSH (a lab workstation or an HPC login node) so Claude can run jobs on it. Use it to connect to a remote workstation with a GPU, or your existing HPC cluster. Claude Science uses your existing `~/.ssh/config`, authenticates with your key or `ssh-agent`, and installs nothing on the host itself.

66

7On Team and Enterprise plans, your organization can turn SSH hosts off. When it has, **Add SSH host** isn't available, hosts you added earlier stay listed but refuse new commands and file transfers, and a job that's already running can still be stopped and its results collected (see [SSH hosts](/docs/claude-science/admin-controls#ssh-hosts) in the admin controls).

7On Team and Enterprise plans, your organization can turn SSH hosts off. When it has, **Add SSH host** isn't available, hosts you added earlier stay listed but refuse new commands and file transfers, and a job that's already running can still be stopped and its results collected (see [SSH hosts](/docs/claude-science/admin-controls.md#ssh-hosts) in the admin controls).

88

99## Adding a host

1010

claude-science/update-claude-science Changed · +2 / -2 lines

from line 50

5050

5151## If the update still fails

5252

53* **On Windows, an update said "restart aborted" or "The update is installed"**: don't quit Claude Science or run `claude-science stop`, and don't shut down or restart your PC, or sign out of Windows. Letting your PC sleep doesn't close Claude Science. If security software is blocking the new version, Claude Science might not open again. Select **Update now** or **Restart to update** again after a few minutes. If it still fails, ask your IT team whether security software is blocking Claude Science, and send them [Where the app stores data](/docs/claude-science/manage-on-devices#where-the-app-stores-data). If you don't have an IT team, contact [Claude support](https://support.claude.com/en/articles/9015913-how-to-get-support).

53* **On Windows, an update said "restart aborted" or "The update is installed"**: don't quit Claude Science or run `claude-science stop`, and don't shut down or restart your PC, or sign out of Windows. Letting your PC sleep doesn't close Claude Science. If security software is blocking the new version, Claude Science might not open again. Select **Update now** or **Restart to update** again after a few minutes. If it still fails, ask your IT team whether security software is blocking Claude Science, and send them [Where the app stores data](/docs/claude-science/manage-on-devices.md#where-the-app-stores-data). If you don't have an IT team, contact [Claude support](https://support.claude.com/en/articles/9015913-how-to-get-support).

5454* **Claude Science still asks you to update**: on Windows, if an update said "restart aborted" or "The update is installed", follow the entry above instead. On a Mac where Claude Science opens in its own window, quit it with Command-Q, wait ten seconds, then open it from your **Applications** folder. Anywhere else, run `claude-science stop` in a terminal (on a Mac or Linux, if the command isn't found, type `~/.local/bin/claude-science stop`), quit Claude Science if you're on a Mac, then open it again. If `claude-science stop` says "Not killing pid", follow [the terminal steps](#from-a-terminal-or-on-a-server).

55* **The message says "restart aborted" or "The update is installed"**: on a Mac or Linux, if you started Claude Science from a terminal, follow [the terminal steps](#from-a-terminal-or-on-a-server). Otherwise, follow [restart aborted](/docs/claude-science/troubleshooting#restart-aborted).

55* **The message says "restart aborted" or "The update is installed"**: on a Mac or Linux, if you started Claude Science from a terminal, follow [the terminal steps](#from-a-terminal-or-on-a-server). Otherwise, follow [restart aborted](/docs/claude-science/troubleshooting.md#restart-aborted).

5656* **The message says "the new version couldn't start"**: you're still on the version you had. Updating again, even from a terminal, gets the same version until a newer one is released, so contact [Claude support](https://support.claude.com/en/articles/9015913-how-to-get-support) with the exact words you see.

5757* **Anything else**: on a Mac where Claude Science opens in its own window, contact [Claude support](https://support.claude.com/en/articles/9015913-how-to-get-support) with your version and the exact words you see. Anywhere else, [install Claude Science again](#install-claude-science-again), and contact support if that doesn't work.

5858

connectors/building/mcp-apps/instance-supersession Changed · +1 / -1 lines

from line 240

240240## Next steps

241241

242242* [Set `ui.domain` for Claude](/docs/connectors/building/mcp-apps/getting-started#set-ui-domain-for-claude): how to compute `_meta.ui.domain` for Claude

243* [Troubleshoot MCP Apps](/docs/connectors/building/mcp-apps/troubleshooting): developer tools and fixes when a widget doesn't render

243* [Troubleshoot MCP Apps](/docs/connectors/building/mcp-apps/troubleshooting.md): developer tools and fixes when a widget doesn't render

244244* [SDK API reference](https://modelcontextprotocol.github.io/ext-apps/api/index.html): `registerAppTool`, `App`, and `McpUiResourceMeta`

245245

connectors/building/mcp-apps/troubleshooting Changed · +2 / -2 lines

from line 133

133133## Next steps

134134

135135* [Set `ui.domain` for Claude](/docs/connectors/building/mcp-apps/getting-started#set-ui-domain-for-claude): compute the sandbox origin Claude expects for your app

136* [Design guidelines](/docs/connectors/building/mcp-apps/design-guidelines#mobile-guidelines): mobile layout, safe areas, and sizing rules that prevent clipped or invisible content

137* [Get started with MCP Apps](/docs/connectors/building/mcp-apps/getting-started#build-your-own-mcp-app): SDK quickstart, examples, and agent skills

136* [Design guidelines](/docs/connectors/building/mcp-apps/design-guidelines.md#mobile-guidelines): mobile layout, safe areas, and sizing rules that prevent clipped or invisible content

137* [Get started with MCP Apps](/docs/connectors/building/mcp-apps/getting-started.md#build-your-own-mcp-app): SDK quickstart, examples, and agent skills

138138

connectors/github/index Changed · +2 / -2 lines

from line 144

144144

145145## Next steps

146146

147* [Add a connector from the directory](/docs/connectors/getting-started#add-a-connector-from-the-directory): find a connector for another service in the directory and connect it

148* [Connectors directory](/docs/connectors/directory): browse verified and community integrations

147* [Add a connector from the directory](/docs/connectors/getting-started.md#add-a-connector-from-the-directory): find a connector for another service in the directory and connect it

148* [Connectors directory](/docs/connectors/directory.md): browse verified and community integrations

149149

government/account/profile Changed · +2 / -2 lines

from line 15

1515| **Email** | The address you sign in with. It is the unique identifier for your account, and notifications such as sign-in links are sent to it. | Your agency's identity provider. |

1616| **Organization** | The organization you belong to within your agency's tenant. A tenant is your agency's overall space in Claude for Government, and it can contain several organizations (for example, one per bureau or office). Your organization determines which administrators manage your access and which usage pool your activity draws from. | Set when your account was created or when directory sync placed you. |

1717| **Role** | What you are allowed to manage. **User** means you can use Claude but do not administer anything. **Owner** means you can manage your organization's users, seats, and settings. **Primary Owner** is the same as Owner with a few additional safeguards: a primary owner cannot be removed by other owners, and each organization can have at most three of them. | Assigned by an organization owner or by directory sync. |

18| **Seat tier** | The allowance you have been given. A seat tier is a named package that sets two things for you: how much Claude usage you get (shown on the [Usage](/docs/government/account/usage) tab) and which Claude models you are allowed to use. The tiers themselves, their names, and what each one includes are defined by your agency, so the name you see here is specific to your deployment. | Assigned to you by an organization owner. |

18| **Seat tier** | The allowance you have been given. A seat tier is a named package that sets two things for you: how much Claude usage you get (shown on the [Usage](/docs/government/account/usage.md) tab) and which Claude models you are allowed to use. The tiers themselves, their names, and what each one includes are defined by your agency, so the name you see here is specific to your deployment. | Assigned to you by an organization owner. |

1919

2020The name and avatar at the top come from your agency's directory. The avatar is generated from your name; you cannot upload a custom picture.

2121

from line 38

3838>

3939> This section only appears when you do not have a seat tier. If a tier name is shown in the Seat tier field, you can skip this section.

4040

41Without a seat tier you can sign in and see the portal, but you cannot send any messages to Claude, and the [Usage](/docs/government/account/usage) tab will show "No seat tier assigned" instead of your allowance. This is the expected state for a brand-new account that has not yet been given a seat, or for an account whose seat was deliberately removed.

41Without a seat tier you can sign in and see the portal, but you cannot send any messages to Claude, and the [Usage](/docs/government/account/usage.md) tab will show "No seat tier assigned" instead of your allowance. This is the expected state for a brand-new account that has not yet been given a seat, or for an account whose seat was deliberately removed.

4242

4343Ask your organization's owner to assign you a seat. Once they do, the tier name appears here immediately and you can start using Claude straight away without signing out and back in.

4444

government/changelog Changed · +4 / -0 lines

from line 18

1818

1919**Claude Code 2.1.302**

2020

212026-10-16

22

2123* Changed unattended runs of the Claude Code command-line tool, such as `claude -p`, where **Auto mode** is set to **Allowed**: on Claude Code 2.1.302 or later they start in Auto mode only when Auto mode is chosen for them.

2224

2325**Claude Code 2.1.299**

26

272026-10-14

2428

2529* Changed **Managed hooks** on Windows devices without Git for Windows: on Claude Code 2.1.299 or later, a managed hook written as a Bash command that checks a tool call or a permission request blocks that action when the command fails.

2630

government/desktop/plugins Changed · +3 / -1 lines

from line 4

44

55> **Who this is for:** Anyone who uses Claude Desktop in Claude for Government and wants to find, install, or create plugins.

66

7A plugin is a package that adds capabilities to Claude in a single step, such as skills, slash commands, sub-agents, and hooks. Plugins work in Cowork and in Code. See the [Plugins overview](/docs/plugins/overview) for more on what a plugin can contain.

7A plugin is a package that adds capabilities to Claude in a single step, such as skills, slash commands, sub-agents, and hooks. Plugins work in Cowork and in Code. See the [Plugins overview](/docs/plugins/overview.md) for more on what a plugin can contain.

88

99<video controls preload="metadata" playsInline src="https://mintcdn.com/claude-ai/gGFKuNSbKYs4JMmK/images/government/videos/user-04-connectors-plugins-and-skills.mp4?fit=max&auto=format&n=gGFKuNSbKYs4JMmK&q=85&s=953d050db0ec38bd9d0db7399e96cf40" aria-label="Video walkthrough: Connectors, plugins, and skills from your agency" />

10

11Video: Connectors, plugins, and skills from your agency (2 min 6 s). Narrated with an AI-generated voice, with on-screen captions.

1012

1113> **Transcript**

1214>

government/tenant-admin/admins Changed · +1 / -1 lines

from line 26

2626

2727## Removing a tenant administrator

2828

29Click **Remove** next to a name to revoke their tenant admin access. The change takes effect immediately. The person keeps their account and whatever organization role they have, but loses access to this portal. Unless they are an organization's [Primary Owner](/docs/government/org-admin/users#understanding-roles), they can also no longer use an emailed sign-in link.

29Click **Remove** next to a name to revoke their tenant admin access. The change takes effect immediately. The person keeps their account and whatever organization role they have, but loses access to this portal. Unless they are an organization's [Primary Owner](/docs/government/org-admin/users.md#understanding-roles), they can also no longer use an emailed sign-in link.

3030

3131> [!WARNING]

3232>

government/tenant-admin/seats Changed · +1 / -1 lines

from line 37

3737

3838## How seats affect new users

3939

40When a new person is placed in an organization (by a routing rule on the [Identity and access](/docs/government/tenant-admin/identity-and-access) page), they're automatically given a seat if one is free. Tiers are filled in a fixed order, so the first tier is filled before the next is started. If every tier the organization has is completely full, or the organization has no seats distributed to it yet, the new person is placed in the organization without a seat. They can sign in and see the portal, but cannot send messages to Claude until someone gives them a seat tier. An organization owner can do that on the organization's Users page, and so can you from the organization's admin view.

40When a new person is placed in an organization (by a routing rule on the [Identity and access](/docs/government/tenant-admin/identity-and-access.md) page), they're automatically given a seat if one is free. Tiers are filled in a fixed order, so the first tier is filled before the next is started. If every tier the organization has is completely full, or the organization has no seats distributed to it yet, the new person is placed in the organization without a seat. They can sign in and see the portal, but cannot send messages to Claude until someone gives them a seat tier. An organization owner can do that on the organization's Users page, and so can you from the organization's admin view.

4141

4242The first seats you give an organization are also used for the people already in it. If nobody in the organization holds a seat when you save its first allocation, members without a seat tier are seated automatically from those seats, Primary Owners first. This covers an organization that was created with only a Primary Owner and given seats afterwards. Anyone the new seats do not cover stays without a seat tier until an owner assigns one. Later changes to an allocation do not repeat this. They add or remove free seats for new arrivals and for the organization's owners to assign, and directory-provisioned users with a group-to-tier mapping are seated from the added seats on the next sync.

4343

government/tenant-admin/tenant-restrictions Changed · +1 / -1 lines

from line 23

2323* Claude Code

2424* The Claude for Microsoft 365 add-ins

2525* Direct API calls

26* SCIM directory provisioning and the [Compliance API](/docs/government/org-admin/compliance-api)

26* SCIM directory provisioning and the [Compliance API](/docs/government/org-admin/compliance-api.md)

2727

2828The same check applies at sign-in, so a user signing in to a tenant that is not on the list sees the refusal at the sign-in screen rather than after authentication completes.

2929

office-agents/dictation Changed · +1 / -1 lines

from line 18

1818> authentication. It is not supported when Claude for M365 connects through

1919> a third-party platform such as Amazon Bedrock, Google Cloud Vertex AI,

2020> Azure AI Foundry, or an LLM gateway. See

21> [Use Claude for M365 with third-party platforms](/docs/office-agents/third-party-platforms)

21> [Use Claude for M365 with third-party platforms](/docs/office-agents/third-party-platforms.md)

2222> for platform support details.

2323

2424## Use dictation

office-agents/opentelemetry Changed · +2 / -2 lines

from line 116

116116Deployments that authenticate against Bedrock, Vertex AI, Foundry, or a

117117gateway supply the same two keys through the customer-configuration

118118channels described in

119[Use Claude for M365 with third-party platforms](/docs/office-agents/third-party-platforms).

119[Use Claude for M365 with third-party platforms](/docs/office-agents/third-party-platforms.md).

120120

121121| Key | Format | Description |

122122| - | - | - |

from line 129

129129then the bootstrap response.

130130

131131**Manifest URL parameter.** Append the keys to the taskpane URL in your

132custom manifest, with `otlp_headers` after `#` so the credential stays out of [server request logs](/docs/office-agents/third-party-platforms#keep-credentials-out-of-the-manifest-url-query-string).

132custom manifest, with `otlp_headers` after `#` so the credential stays out of [server request logs](/docs/office-agents/third-party-platforms.md#keep-credentials-out-of-the-manifest-url-query-string).

133133

134134```text theme={null}

135135https://pivot.claude.ai/taskpane.html?otlp_endpoint=https://otel-collector.example.com#otlp_headers=Authorization=Bearer%20<token>

office-agents/performance Changed · +2 / -2 lines

from line 86

8686* Close the files you are not working in.

8787* For large files, work in one file at a time.

8888* When Claude works across Office apps, ask it to finish in one app before it

89 moves to the next. See [Work across M365 apps](/docs/office-agents/work-across-apps).

89 moves to the next. See [Work across M365 apps](/docs/office-agents/work-across-apps.md).

9090* If Excel runs out of memory with several workbooks open, open Excel in a new

9191 instance.<sup id="cite-ref-8-b">[\[8\]](#cite-note-8)</sup>

9292

from line 272

272272

273273Claude sends one trace for every user turn, including steps that failed.

274274Search your collector for traces at the time of the problem. See

275[Configure a custom OpenTelemetry collector](/docs/office-agents/opentelemetry).

275[Configure a custom OpenTelemetry collector](/docs/office-agents/opentelemetry.md).

276276

277277## References

278278

third-party/claude-desktop/claude-api Changed · +2 / -2 lines

from line 10

1010

1111## Choose an authentication approach

1212

13There are three options. With neither a static key nor a credential helper configured, each user sees **Sign in with Claude Console** on first launch: the app opens the browser, the user signs in and selects a Claude Console (API) organization, and the app keeps that sign-in encrypted on the device and renews it as needed, without creating an API key. Usage is billed to that Console organization, and about a month after a user signs in, the app asks them to sign in again. Alternatively, place a static API key in the managed configuration as `inferenceAnthropicApiKey`, or, where static keys aren't permitted, set [`inferenceCredentialHelper`](/docs/third-party/claude-desktop/configuration#inferencecredentialhelper) to an executable that fetches a short-lived credential at runtime; see [Write a credential helper](/docs/third-party/claude-desktop/credential-helper). Browser sign-in reaches `platform.claude.com`, `hcaptcha.com`, and `*.hcaptcha.com` in addition to `api.anthropic.com`.

13There are three options. With neither a static key nor a credential helper configured, each user sees **Sign in with Claude Console** on first launch: the app opens the browser, the user signs in and selects a Claude Console (API) organization, and the app keeps that sign-in encrypted on the device and renews it as needed, without creating an API key. Usage is billed to that Console organization, and about a month after a user signs in, the app asks them to sign in again. Alternatively, place a static API key in the managed configuration as `inferenceAnthropicApiKey`, or, where static keys aren't permitted, set [`inferenceCredentialHelper`](/docs/third-party/claude-desktop/configuration#inferencecredentialhelper) to an executable that fetches a short-lived credential at runtime; see [Write a credential helper](/docs/third-party/claude-desktop/credential-helper.md). Browser sign-in reaches `platform.claude.com`, `hcaptcha.com`, and `*.hcaptcha.com` in addition to `api.anthropic.com`.

1414

1515Claude Desktop versions before 3.1623.0 created a personal API key in the Console organization when a user signed in with Claude Console, and the app doesn't delete those keys. A device that signed in that way and then updates to 3.1623.0 or later keeps using its key until you revoke the key in Claude Console or the user signs out. The device's next sign-in creates no key.

1616

17[Code](/docs/third-party/claude-desktop/code) sessions also ask Anthropic for the Claude Code [server-managed settings](https://code.claude.com/docs/en/server-managed-settings) of the organization that owns the credential. See [Interaction with Claude Code's own managed settings](/docs/third-party/claude-desktop/code#interaction-with-claude-code%E2%80%99s-own-managed-settings) for how they combine with your Claude Desktop configuration.

17[Code](/docs/third-party/claude-desktop/code.md) sessions also ask Anthropic for the Claude Code [server-managed settings](https://code.claude.com/docs/en/server-managed-settings) of the organization that owns the credential. See [Interaction with Claude Code's own managed settings](/docs/third-party/claude-desktop/code.md#interaction-with-claude-code%E2%80%99s-own-managed-settings) for how they combine with your Claude Desktop configuration.

1818

1919## Configure the app

2020

One read of Claude Documentationclaude-docs-20261022T180711Z

What this read moved

claude-science/example-gallery/soc-schelling-small Changed · +1 / -1 lines

claude-science/mac-app-window Changed · +2 / -2 lines

claude-science/monitor-usage Changed · +1 / -1 lines

claude-science/remote-compute-clusters Changed · +1 / -1 lines

claude-science/update-claude-science Changed · +2 / -2 lines

connectors/building/mcp-apps/instance-supersession Changed · +1 / -1 lines

connectors/building/mcp-apps/troubleshooting Changed · +2 / -2 lines

connectors/github/index Changed · +2 / -2 lines

government/account/profile Changed · +2 / -2 lines

government/changelog Changed · +4 / -0 lines

government/desktop/plugins Changed · +3 / -1 lines

government/tenant-admin/admins Changed · +1 / -1 lines

government/tenant-admin/seats Changed · +1 / -1 lines

government/tenant-admin/tenant-restrictions Changed · +1 / -1 lines

office-agents/dictation Changed · +1 / -1 lines

office-agents/opentelemetry Changed · +2 / -2 lines

office-agents/performance Changed · +2 / -2 lines

third-party/claude-desktop/claude-api Changed · +2 / -2 lines