Hand off and ask

Two sessions often need each other. One has read the whole subsystem; the other is about to change it. Offsite has two moves for that, both on the card’s More Actions menu.

0:55

A plan worked out with Claude goes to a new Codex session for a second opinion, with the model and effort you pick: it reads the discussion itself and comes back with its review. Later, Claude asks an older Grok session how it set up the webhooks and answers in its own card. A 55-second film, with sound.

Ask another session

Ask another session… gets you context from another agent session without leaving the card you are on. Pick the session, type your question or leave it empty for a summary, and the app puts a prompt into this card’s session. Its agent reads the other session’s transcript with offsite agents context and answers right here.

Use it for “what did the other agent conclude about the migration?” without switching cards, re-reading a long scrollback, or interrupting anything. The other session is not touched; it does not even know it was read.

You can ask the running agent sessions of the same project: this workspace and the worktrees of its repository. A Codex session can be asked once its first turn has given Offsite its CLI session ID. With no session to ask, the entry is greyed out.

Hand off to a new session

Hand off to a new session… works the other way round: it starts a fresh session that pulls the current session’s context and continues from there.

Pick the agent that takes over, Claude Code, Codex or Grok, with its model, its effort and YOLO if you want it. Then write what it should do, or leave the message empty to continue the work. The new session starts in the same workspace, loads this session’s context first and then follows your message.

Use it when a conversation has grown long, or when the next phase of work deserves a clean slate that still knows what happened.

Why handoffs go to a new session

Offsite deliberately cannot deliver a message into an existing session.

That capability existed and was removed in August 2026, for a concrete reason: delivering text into a running agent means typing into its terminal, and a terminal is a shared channel. A delivery arriving while the user had a half-written prompt in the same input merged with that draft and was submitted as one prompt, with the user’s authority. No amount of care at the sending end fixes that; the channel itself is the problem.

So the rule is now structural: a session may read another session, never write into one. Handoffs create a new session that pulls context, which has no shared input to collide with.

Text delivered into a running session shares the input with your draft, so a session only reads another and a handoff starts a new one.

For agents

An agent running inside an Offsite session can do the read side itself:

offsite agents list       # the user's other agent sessions, with activity
offsite agents context    # another session's transcript, normalized

There is no send. See Offsite for agents.

Ask an agent for a handoff (“carry on in a new session”) and it starts the new session itself, with the same pull of context; see Agents that steer Offsite.

Details

• agents list reports each visible agent session with an activity of busy, attention or idle, derived from the same turn state that drives notifications.
• Session selectors resolve by id, id prefix, then name; running sessions are matched before stopped ones so a dead session cannot shadow a live one.
• Transcript reads are capped (4 MB tail, blocks trimmed to 2000 characters, tool output to 300) so a huge conversation cannot blow up the caller’s context.
• Reads act as the calling session’s agent, identified by OFFSITE_SESSION_ID — never as one of your paired devices. That keeps a read from being mistaken for you looking at a card, which matters for attention handling.