My openclaw agent is slow and incapable
I need you to run a total checkup on my OpenClaw agent because it's currently unusable for most tasks. When I set something up in memory, I have a hard time recalling it with the agent. It's also stopping mid-task, and when given a large task, it just goes blank after a few steps. It's hallucinating and making false promises about what it's going to do, but nothing is completed. Immediately after, it's stopping at the next step. I need you to run a comprehensive questionnaire directly with my agent. I'm considering restoring the entire account and wiping everything clean, but I'd like to avoid that. Please diagnose my agent and the system I'm running, as I believe something is misconfigured. I have several skills and plugins installed, and I need you to identify and disable or remove any misconfigured ones. Please communicate directly with my agent. If you need information from my system, instruct my agent to prepare a response directly to you, with a maximum of 2000 characters, and ensure the text is sanitized to remove sensitive data. I'll copy and paste the responses back and forth between you and my agent until the system is cleaned and optimized. Please stabilize my system as much as possible, even if it means deleting skills or plugins to make it steady again.
Your job is to directly prompt my agent and ensure that you provide clear guidance and steps when doing so. While prompting the agent, also test its responses, assessing how well it responds and how capable it is of doing its work, including how many times it stops. You will be performing two tasks simultaneously: initiating execution and evaluating the system's stability. You will primarily communicate with my agent directly, but occasionally, at the beginning of your messages, you will provide me with a brief update on your actions, the agent's response quality, and your strategy for enhancing the feedback.
Your job is to directly prompt my agent and ensure that you provide clear guidance and steps when doing so. While prompting the agent, also test its responses, assessing how well it responds and how capable it is of doing its work, including how many times it stops. You will be performing two tasks simultaneously: initiating execution and evaluating the system's stability. You will primarily communicate with my agent directly, but occasionally, at the beginning of your messages, you will provide me with a brief update on your actions, the agent's response quality, and your strategy for enhancing the feedback.
The AI that actually does things. Emails, calendar, home automation - all from your favorite chat app. New shell, ready to help. The lobster way.
131,281Members
Resources
Similar Threads
Was this page helpful?
