Chat with an Agent

Agent Chat is a standalone conversation with an Agent or Crew. Use it when you want to ask a question, investigate a topic, shape a plan, or start work without first opening a task.

Each conversation is backed by its own chat session and chat messages. The session belongs to the person who created it. Other organization members do not receive access to that session through the current chat APIs.

Agent Chat and Task Conversation are separate

Agent Chat and Task Conversation can both start Agent runs, but they store different records:

Agent Chat uses standalone chat sessions and chat messages.

Task Conversation uses comments on a task and routes default Agent responses through the task's Agent Assignee.

Sending a message in Agent Chat does not create a task comment. Sending a comment in Task Conversation does not add a message to an Agent Chat session. The two surfaces share execution infrastructure, but their conversation histories remain separate.

Use Agent Chat for open-ended work. Use a task when the work needs an Assignee, status, review, acceptance criteria, or a durable task record.

Start and resume a session

Open an Agent's chat view, then send a message or attachment. A new session is created when the first message is sent.

When you return to an Agent with no session explicitly selected, Sharkly opens that Agent's most recent active standalone session that contains messages. Choose New session when you want a clean conversation instead.

The session picker lets you:

open another active session for the same Agent or Crew;

rename an active session;

archive an active session;

open an archived session from history.

Archived sessions are read-only. Their messages remain available, but you cannot send another turn in them.

Choose a chat mode

The selected mode is stored with the session. New chats start in Execute unless you choose another default.

Mode	Behavior
Explore	Read-only investigation. Ask the Agent to inspect, explain, compare, or summarize without making changes.
Plan	The Agent can propose actions for your review. Confirm a proposal before Sharkly creates the planned task and subtasks; you can confirm without starting or confirm and start the work.
Execute	The Agent can act directly through its configured CLI and return the result in the session.

You can change the mode before a turn. Keep Explore selected when you want analysis only, and review Plan proposals before confirming them.

Send messages and files

The composer accepts text and file attachments. A message can contain only an attachment when no text is needed.

Agent output streams into the session while a run is active. You can stop the active run. If you send another message while the Agent is already working, Sharkly keeps the message in the session and queues it for the next turn instead of starting a parallel duplicate run for the same Agent.

Unsent draft text is retained per Agent or Crew conversation when you switch between chats in the current browser.

Agent access and availability

You can start a session only with an Agent or Crew you are allowed to use. The target must still be active and available when you send each turn.

If an Agent is archived or is no longer accessible to you, the existing session history remains readable but the composer is disabled. Start a new chat with an accessible Agent to continue.

Move from chat to tracked work

When a discussion produces a concrete action:

Summarize the goal, scope, constraints, and acceptance criteria.

In Plan mode, review and edit the proposed task and subtasks before confirming them, or create a task yourself.

Assign the task to the appropriate person, Agent, or Crew.

Continue task-specific feedback in Task Conversation or Activity so the discussion is stored as task comments.

The original Agent Chat session remains separate history. Copy any context that the task must retain into its description or comments.

Agent Chat and Task Conversation are separate#

Start and resume a session#

Choose a chat mode#

Send messages and files#