Many conversations at once
When a flow drives several agents, each conversation they hold gets a transcript of its own, and each outworlder gets one of what it asks you. No transcript holds every agent's lines at once. A run lands on the agent of the first conversation it opens; press shift+tab to read the others one at a time, and ← for the monitor, where the whole run is watched.
At a glance
- You will switch between the transcripts of the agents and conversations of a run, and read the marks that say who is working and who has news.
- Use it when a flow drives two or more agents, or one agent opens many conversations.
- You need a flow running in
hmz. A flow with two agents, such asrlar, shows it best.
Try it
While a flow with several agents runs, press shift+tab with nothing typed. Press it again to step on, and tab to step back.
How it works
A conversation (a session) is one exchange with one agent kept across its turns. A flow decides how many each agent holds: a reviewer that starts fresh every round opens a new one each time, and a fan-out holds several at once. Every conversation gets a transcript of its own, and so does each agent, with all of its conversations running down it.
When a run starts, the screen lands on the agent of the first conversation it opens, with what you started it on at the top, and follows that agent into its later conversations. Before any conversation opens, the screen shows only the interface's own lines.
What you are reading decides more than what is on the screen:
- The transcript, drawn from the top with a line saying whose it is.
- Where a line you type goes. See Talking to a running turn.
- Who answers
/btw: a side copy of that conversation, or the btw agent on the monitor. - What
/clearclears: only that transcript.
Example: follow a builder and a tester
A project flow runs a builder and a tester side by side. Press the keys to step through the transcripts, as shift+tab does:
Where a run lands: the agent of the first conversation it opens, drawn from the top. unread marks the other agent: it has said something you have not seen.
What to look at, by number:
─ reading builder ─. You are on the builder's transcript, where the run landed, drawn from the top so it reads as a whole.- One line per agent above the editor. The name the flow gives it, what it runs as
cli/model:effort, and●while it has a turn open. These stay whatever you read.· readingmarks the agent whose transcript is on the screen. · unreadmarks an agent that has said something since you last read it: the tester has been working while you read the builder.─ reading tester · conversation 1 ─. One conversation's own transcript.
Check it worked
- The line at the top of the transcript names the conversation you meant to read.
- The agent you are reading carries
· readingabove the editor. - A line you type now goes to that conversation, or to the open turn of the agent you read.
The keys
| Key | Reads |
|---|---|
| shift+tab | The next conversation that is running, then each outworlder of the flow, then round again. |
| tab | The one before. |
| ←, then enter on a node | Any agent or conversation, working or not, picked on the monitor. |
shift+tab steps only between conversations that are running, so with ten agents it skips the ones that are idle. Once you are reading a conversation, you stay on it after its turn ends, until you press a key. To reach one that has ended, or an agent that has not started yet, use the monitor.
An outworlder's transcript holds only what that outworlder puts to you, and a line typed there answers it. See Questions.
What the lines above the editor say
Each line is one agent: the name the flow gives it, what it runs as cli/model:effort, and then what it is holding.
| Mark | Means |
|---|---|
● 1 | It has a turn open. The number is how many conversations it holds. |
○ 2 | It is between turns, and holds two conversations. |
reading | Its transcript is the one on the screen. |
unread | It has said something since you last read it. |
One agent, many conversations
An agent can hold many conversations. A Ralph loop opens a fresh one every round, and a fan-out holds several at once. Each has a transcript of its own, numbered in the order the agent opened them, and they all run down that agent's one transcript too, which is never wiped when a new one opens. When an agent holds more than one, each turn says which:
● agent is working · conversation 2 of 2How many conversations each agent holds is up to the flow. See Many turns at once and Branching a conversation.
How much the screen keeps
The last 32 transcripts, a conversation's going before an agent's, and the last 2,000 lines of each. Anything older is gone from the screen but not from the run's trace.
Troubleshooting
shift+tab skips the agent I want
It steps only between conversations with a turn open. Press ← for the monitor, walk to the agent or session, and press enter: that reads it whether it is working or not.
The transcript I was reading stopped moving
You are still on that conversation, and its turn has ended. A Ralph loop's next round is a new conversation. Press shift+tab to go on to the one that is running, or read the agent itself from the monitor, which follows it into each new conversation.
A line I typed is still on the pin
A line typed on one conversation's transcript waits for that conversation's next turn, which a loop that starts fresh every round never takes. Read the agent itself from the monitor, which says to whichever of its conversations has a turn open, and type it again. See Talking to a running turn.
Next steps
- Watching a run: every agent and conversation on one screen, to pick from.
- Talking to a running turn: what a line typed here does.
- Exporting a run: package every conversation a run opened.