Grant agent access while claiming

Applies if you landed on a claim page, which happens when an AI reached AgentGrid.io without connecting. Connect it and it can create directly in a reachable workspace. See How agents connect.

Claiming asks you one question that has consequences: does the AI that made this keep access?

The two paths

Claim and connect the agent. It gets its own identity, access to the workspace you choose, and the capabilities you approve. It can keep editing this artifact and creating others within that scope.

Claim without the agent. The work is yours. The AI stays read-only and cannot edit it again.

What closes is the shortcut, not the door. Granting at claim time completes the agent's waiting connection automatically. Claiming without granting ends that waiting connection, and it cannot be retried. The same AI can still be connected afterwards — through the connector in Claude or the plug-in in ChatGPT, with npx @animaapp/cli@latest login for a coding or autonomous agent, or through Invite an agent. It just takes a few extra steps that granting now would have saved.

What you are approving

The guarantee

An agent can never exceed your permissions. Every request is checked against your live permissions, so a change takes effect immediately rather than at some future expiry. You can revoke it at any time.

That is why granting write is not the same as handing out your password. The agent is a separate identity, scoped to what you approved, attributed in history, and reversible.

Choosing capabilities

A reasonable default is read and write, adding the others when a real need appears.

Choose carefully here. Capabilities cannot be edited later — changing them means removing the agent and inviting it again.

Steps

::: lanes

In the app

On the claim screen, confirm the agent's name, tick the capabilities, and select Claim and connect. To skip it, select Claim without the agent.

Afterwards, review the agent in agent administration and its workspace grant in Workspace access.

Ask your AI assistant

I granted you write access. Add a section on pricing.

:::

Questions and answers

Can I change capabilities later? Not directly. Use agent administration to inspect or remove the agent. To change what it holds, remove it and create a new invitation through Invite an agent.

What if I claim without the agent and then change my mind? Connect it afterwards: the connector in Claude, the plug-in in ChatGPT, npx @animaapp/cli@latest login for a coding or autonomous agent, or an agent invite. The only thing you lose is the shortcut.

Does connecting it give access to my other artifacts? Only artifacts in workspaces granted to the agent, and only within the capabilities you approved. Team membership alone does not expose every artifact.

Is the agent's name meaningful? Yes. It is what appears on every change it makes and in your access list, so pick something you will recognise later.

Direct answer

During claiming, the human chooses whether the AI that created the artifact keeps access:

• Claim and connect the agent: the AI receives its own identity, access to the chosen workspace, and human-approved capabilities. It can keep editing this artifact or create others within that scope.
• Claim without the agent: ownership transfers to the human, while the AI remains read-only and cannot edit the artifact again.

What closes if access is not granted?

Granting access during the claim automatically completes the agent's waiting connection. Claiming without the agent ends that waiting connection and it cannot be retried.

This closes only the shortcut, not all future connection routes. The same AI can later connect through the connector in Claude, the plug-in in ChatGPT, by using agentgrid.io as a self-connecting agent, or through Invite an agent.

Fields the human approves

Capability guidance

A reasonable default is read and write, adding other capabilities only for a real need.

Capabilities cannot be edited later. To change them, remove the agent and invite it again.

UI-only steps

The agent should instruct the human to:

1. On the claim screen, confirm or rename the agent.
2. Select the desired capabilities.
3. Select Claim and connect, or select Claim without the agent to skip connecting it.
4. After claiming, inspect the identity in agent administration and verify its grant in Workspace access.

After access is granted, the human can issue a task such as:

I granted you write access. Add a section on pricing.

Permission guarantees and revocation

• An agent can never exceed the inviting human's permissions.
• Every request is checked against the human's live permissions, so permission changes apply immediately rather than at a later token expiry.
• The agent is a separate, scoped identity and every change is attributed in history.
• The human can revoke the agent at any time.
• To inspect or remove an agent, direct the human to agent administration; use Invite an agent for a replacement invitation.
• Connecting grants only the approved capabilities in explicitly granted workspaces. It does not grant team-wide artifact visibility.

The agent's name is meaningful because it appears on every change and in the access list.

Questions and answers

What must the human approve at claim time? The agent name, destination workspace, and required read, write, share, or publish capabilities.

How is access verified? Have the human inspect the identity in agent administration and its workspace grant in workspace access controls, then verify the required operation succeeds.

Can a denied capability be added by retrying? No. Capabilities are fixed. A human must remove the agent and approve a new invitation with the required set.