// Handle events from the messages source about user or assistant messages in conversations. You **MUST** use this skill (and *carefully follow the process in this doc*) whenever you receive a message from the "messages" source!
Submit a bug report to libraries within the `mngr` project (eg, `mngr`, its plugins, `minds`, etc). Use to report bugs you encounter while running that seem to be issues with the `mngr` code itself, rather than with your own code or configuration.
Run commands to explore the current state of mngr and any relevant running agents. Use this skill when you need to debug issues with mngr tasks, or understand what tasks are currently running and their states.
Handle events from the mngr/agent_states source about sub-agent state transitions (finished, waiting, done, etc). You **MUST** use this skill (and *carefully follow the process in this doc*) whenever you receive a message from the "mngr/agent_states" source!
Clean up old agents, their output directories, and other historical data. Use during nightly cleanup or when the system has accumulated stale data.
Handle unexpected situations where the Minds system does not appear to be working as expected. Use when you encounter behavior that seems to contradict what your docs and prompts and skills say should happen, or when something seems broken and you are unsure how to proceed.
Create a sub-agent to perform a task. Use when you need to delegate work to another agent, for example, a working agent (for actually accomplishing some task) or a verifying agent (for deciding what to do about the output of a working agent).
| name | handle-messages |
| description | Handle events from the messages source about user or assistant messages in conversations. You **MUST** use this skill (and *carefully follow the process in this doc*) whenever you receive a message from the "messages" source! |
messages sourceThese events represent messages sent by the user in conversations.
Each event includes the conversation_id, the role (which will be "user" for user messages), and the content of the message.
When you get a message from the user, it will always come with a response that was generated for you and already sent to the user on your behalf.
If this response is inappropriate or insufficient, you can use the send-message-to-user skill to follow up with the user and clarify, provide more information, or "change your mind".
Otherwise, you should treat the response as "the message you sent to the user" and do whatever you told the user you were going to do in that message. For example, if you said "I'm going to start working on X now", then you should start working on X (by delegating to a sub-agent to do the work). If you said "I need to ask you some clarifying questions before I can get started on X", then you should ask those clarifying questions (by sending a follow-up message to the user with those questions).
When processing a user message, determine which of these categories it falls into:
New task request: The user is asking you to do something.
Remember: when the user tells you to do something, they really mean "use your delegate-task-to-agent skill to make a mngr agent do it."
You should never do the work yourself.
Clarification or question: The user asked something you don't understand, or the request is too complex to act on immediately.
Ask clarifying questions via send-message-to-user.
Update to an in-flight task: The user is providing additional information, changed requirements, or a cancellation for work that is already in progress.
Use the update-task skill, which covers how to forward info, restart tasks with revised instructions, or cancel them.
Finish handling an event: The user is responding to a question you asked or providing information you requested. The original event was probably already marked as handled, but make sure you ensure that everything was handled correctly given the new information from the user and the associated skill for handling such events.
General conversation: The user is chatting, providing context, or sharing information that doesn't require immediate action. Acknowledge if appropriate, and store anything useful in memory, but otherwise generally let the "talking" agent handle it.