DEV Community: Berkant The latest articles on DEV Community by Berkant (@berkantay). https://dev.to/berkantay https://media2.dev.to/dynamic/image/width=90,height=90,fit=cover,gravity=auto,format=auto/https:%2F%2Fdev-to-uploads.s3.us-east-2.amazonaws.com%2Fuploads%2Fuser%2Fprofile_image%2F1446787%2Fa8b35218-e1d5-4f18-9c31-ce2338f85f54.jpeg DEV Community: Berkant https://dev.to/berkantay en I built Lucius to make coding-agent sessions easier to understand Berkant Wed, 29 Jul 2026 20:17:17 +0000 https://dev.to/berkantay/i-built-lucius-to-make-coding-agent-sessions-easier-to-understand-4677 https://dev.to/berkantay/i-built-lucius-to-make-coding-agent-sessions-easier-to-understand-4677 <p>Claude Code and Codex are powerful, but longer sessions can become surprisingly difficult to follow.</p> <p>The terminal quickly fills with:</p> <ul> <li>tool calls</li> <li>shell commands</li> <li>retries</li> <li>errors</li> <li>test output</li> <li>file changes</li> <li>long explanations</li> </ul> <p>By the end of a session, the important questions are often still unclear:</p> <p>What actually changed? Why did it change? What failed? What should I review?</p> <p>That is why I built Lucius.</p> <p>Lucius is an open-source tool for exploring coding-agent sessions in a clearer, more structured way. Instead of scrolling through a long terminal transcript, you can inspect the session and understand the agent’s work more quickly.</p> <p>The goal is not to hide the raw output or blindly summarize everything. It is to make it easier to move between the high-level result and the underlying details.</p> <p>I am especially interested in improving the experience around:</p> <ul> <li>understanding the sequence of actions</li> <li>seeing which files were touched</li> <li>spotting failures and retries</li> <li>reviewing important decisions</li> <li>identifying what still needs human attention</li> </ul> <p>Lucius is still early, so feedback from people who regularly use Claude Code, Codex, or similar coding agents would be genuinely useful.</p> <p>How do you review longer agent sessions today? Do you read the full terminal output, inspect only the final diff, or use another tool?</p> <p>GitHub: <a href="https://github.com/berkantay/lucius" rel="noopener noreferrer">https://github.com/berkantay/lucius</a></p> <p>Feedback, issues, contributions, and stars are appreciated.</p> agents opensource showdev tools