Agent Settings

Last updated: October 2026

Use Settings → Agents to install and update the coding agents that work on your tasks, sign in to them, choose how each one starts, and give them more tools.

At the top of the page, a line names the agent new tasks use: New tasks use {agent}. or No default agent for new tasks. To change it, click Change in Tasks, which opens Default agent.

To choose the agent Ask Kepler uses, set the Ask Kepler uses row. This Ask Kepler default agent is separate from the default for tasks. Its help text reads Used when Ask Kepler has no earlier pick to go on and whenever you choose Use default agent there. The row lists only the agents that can answer in the Ask Kepler panel, which are the agents that run in Chat. If the agent you picked stops being available, the row reads Pick another agent for Ask Kepler: {name} is not available. See Ask Kepler.

Your agents

Kepler supports Claude Code, Codex, GitHub Copilot, OpenCode, Pi, Auggie, Antigravity, Cursor, and Grok Build. Settings lists the agents installed on this machine first, then the rest, each group in that order.

In the section header, Update all runs each installed agent’s own updater, and Re-scan looks again for the agent CLIs installed on this machine. If you install or remove an agent outside Kepler, click Re-scan.

Each agent has a card. The card’s title line shows the agent’s name and these controls:

When the version is unknown, Kepler turns off features that depend on the version, such as sending a new session’s first message automatically. Reinstalling or updating the agent usually fixes this.

If the installed version of an agent, or of one of its components, has a known problem, a warning note appears on the card under the title line. The note names the version and says what to do. Kepler can show these notes:

The card body holds the rest:

Under Program and data folder:

Codex runs your own CLI. Like every other agent, Kepler finds and runs the codex you have installed. For usage figures, Kepler reads the OAuth token from ~/.codex/auth.json, or from CODEX_HOME/auth.json when you have set a custom data folder.

OpenCode needs version 2. Kepler reads and resumes only OpenCode 2’s sessions. A session does not start on an OpenCode 1 binary, and the card shows OpenCode {version} is not supported. If you have both versions installed, point Binary at OpenCode 2. A conversation you started under OpenCode 1 resumes from OpenCode 2’s copy of it.

Antigravity runs in Chat through its Agent Client Protocol (ACP) server. That server is a separate component from its agy terminal CLI. See Agent Integrations.

Use several accounts for one agent

With Accounts, you can run more than 1 login for an agent side by side. Each account has its own credentials and history, but shares your skills, agents, commands, and settings.

To add an account, click Add account, then choose whether its conversations Keep separate or Share with the CLI.

Each account row has the following controls:

• A color picker that tints the agent icon wherever that account runs. The default account stays uncolored, so its picker shows the ⊘ No color mark.
• Sign in.
• An account menu with Rename, Share history with the CLI, Sign in another way, Sign out, and Remove account. On a signed-in Claude Code account, the menu also has Switch account…, which signs that account in with a different Claude login. See Switch the account a Claude Code login uses.

Removing an account stops its running sessions and deletes the sessions stored only in that account. To keep those sessions, click Move to CLI account first. See Agent Sessions.

When an added account stops sharing your settings

An added account shares your own settings, instructions, skills, agents, commands, and other configuration by linking to your files. If the agent saves a change while signed in to that account, the agent can write its own copy in place of the link. The account then stops following your file. Kepler shows this under the account’s row: This account stopped sharing your {items}. Expand the message to see both paths, This account’s copy and Your {item}, and choose one of the following:

• Share your {item} again asks first, then links the account back to your file. Its own copy is saved as a backup next to it, so nothing is lost.
• Keep its own {item} keeps the account’s copy. The row’s menu then offers Share your {item} again if you change your mind.

Switch the account a Claude Code login uses

Switch account… opens Switch the account for {name}?, which names the login now signed in. If your browser is already signed in, choose a different account on the sign-in page. The dialog also warns you about the following:

• How many running sessions use the account, and when they pick up the new login: when they restart, or on their next request.
• On the default account, the one the CLI uses, the switch also changes the account the claude command uses in your own terminals.

Continue opens the usual sign-in. When the sign-in succeeds, Kepler confirms: {name} is now signed in as {login}.

Default mode for new sessions

Every agent that can run in both Chat and Terminal has this setting:

Kepler saves the choice per agent, so one agent can default to Terminal while another stays on Chat. The setting is only a default. As its help text says, You can pick the other one when you start a session. A terminal-only agent shows Terminal with no switch. See Agent Sessions.

Artifacts in new sessions

The following setting is on Claude Code’s card:

A running Chat session has its own Artifacts switch in its agent settings. When you turn it on, Kepler asks Turn on Artifacts?, then restarts the session and resumes the conversation. Once a conversation has artifacts on, they stay on for that conversation. See Agent Sessions.

Let artifact comments start turns

The following setting is on Claude Code’s card:

Running sessions keep the value they started with. When you change the setting, the card lists the sessions still on the previous value. It marks the sessions that Published an artifact, have a Turn in progress, or have Background work running. To apply the new value, click Restart all or Restart selected ({count}). A session that is mid-turn or has background work starts unticked.

Restarting keeps each conversation. It interrupts a turn in progress and stops background work, and a restarted session stops watching the artifacts it published before.

Custom agents

To run an agent that Kepler does not list, add it as a custom agent by giving Kepler its command. To add one, click New custom agent in the section header. To edit one, click its row. Either way, the agent opens as a sub-page with these fields:

Custom agents appear with the built-in agents wherever you pick an agent. With none added, the section reads No custom agents yet.

Kepler encrypts environment values at rest and hides existing values when you edit an agent. To keep a stored secret, leave its value empty (KEY=). To replace it, type a new value.

Save waits until Kepler has stored the agent. If the save fails, the form stays open with the reason: Unable to save {name}: {reason}. Delete asks first, with Delete custom agent?: Deletes {name} and the environment values stored with it. You cannot undo a delete.

MCP servers

Use MCP servers to give your agents extra tools, like a database, a browser, or your issue tracker. The servers here are each agent’s own defaults, and every repository inherits them.

To see the servers an agent loads, pick the agent with Agent. If that agent has more than 1 account, pick an account too. Kepler lists only the agents that expose their MCP servers to Kepler. In the list:

• Click a server to open its own page. There you can turn the server or a single tool off, sign in to a server that needs it, use Also add to… to copy the server to another agent, and Remove server.
• Click Add server in the section header to add one.

A server or tool you turn off here is off in every repository, unless a repository turns it back on for itself. A repository’s own MCP servers section is on its Edit repo page. See MCP Servers.

Below the list, one more setting applies across every agent:

This setting affects the MCP clients on your machine. Separately, Kepler attaches its own workspace MCP server to every session it starts. The workspace MCP server is always on and has no setting. Its tools are named for what they act on, such as task_get, session_start, or worktree_create. They let the agent:

• Read a task and its resources, find tasks, create a task, change its Progress stage, name, or hold, and archive it.
• Read, write, and delete the task’s notes.
• Attach and detach issue, pull request, and URL links, and attach or detach files.
• List the agents, repositories, and folders in your workspace. An agent can name a repository by one of its aliases.
• Create, attach, detach, update, or delete one of the task’s worktrees.
• List and read sessions, start a session with the agent it chooses, send follow-ups, rename or archive a session, and read the reports its own sessions send back.
• List pending approvals.
• Read and save themes. See Theme.
• Rebase, merge, and compose commits, when Agent tools turns them on.

Every workspace tool runs without the agent’s own permission prompt, in both Chat and Terminal. Instead, an action that needs your approval, such as deleting a worktree or changing another task, waits for you on an approval card. See Approvals, and Tasks and Resources for the full list of tools.

Agent tools

Use Agent tools to let your agents call Kepler’s own tools for rebasing, merging, and cleaning up commits:

AI Sync and AI Compose add their tools to the workspace MCP server. You need a paid GitKraken subscription to turn either one on. On the Community plan, when you are signed out, or with no organization, a setting that is off shows a lock in place of its checkbox: Not available on the Community plan. Upgrade to unlock. A setting that is already on keeps its checkbox, so you can still turn it off.

When you turn either setting on, Kepler confirms that New agent sessions will pick up this change. Sessions already running keep the tools they started with.

Approvals

Some actions your agents take wait for your approval. Use Approvals to manage the rules that approve those actions for you. An action that no rule covers still asks.

When an agent uses Kepler’s workspace tools to do something that needs your approval, Kepler pins an approval card above that session’s message box. Examples include deleting a worktree, sending a prompt to another session, or rewriting a branch. The card’s Approve menu offers Always approve for that kind of action. Kepler does not offer it for actions that can lose work, or for a setting change that Ask Kepler asks to make. See Agent Sessions.

Always approved lists each rule you create this way. A rule’s row says when and where you added it, Added {date} from {session} on {task}., and how often it has run, Used {count} times, last {when}. or Not used yet. Remove deletes the rule, so that kind of action asks again. With no rules, the section reads No rules yet. When an agent asks, choose Always approve on its request to stop being asked about actions like it.

Usage limits

These settings control how Kepler handles your agent plans’ usage limits: