Running an agent with OpenCode
I have created an agent in OpenClaw that is called the Architect. It is a persistent agent. I have created a skill so that it can talk with my Open Code installation (https://opencode.ai/). The skill allows my agent to create the tasks and ask the open code via CLI to code the various tasks. It is doing it correctly but each time a task is finished, the agent The Architect never gets back to me in telegram and I have to ask him to continue to the next task. I would prefer to have a real autonomy from my agent The Architect that when an open code command is finished that it sends the new command for the next task and tells me the status update. I have tried many times to modify the files AGENT.md but the agent never comes back to me and always answer me if I ask him what the status is.
The AI that actually does things. Emails, calendar, home automation - all from your favorite chat app. New shell, ready to help. The lobster way.
134,827Members
Resources
Recent Announcements
Similar Threads
Was this page helpful?
