Dataclaw Zhiyaowang: 73 Agentic Coding Conversation Logs from CLI Tools

73 coding agent conversation sessions exported from tools like Claude Code and Gemini CLI by zhiyaowang in February 2026. The data captures interactions between human users and agentic coding assistants involving tool-use and text generation sequences.

Use Cases

• Fine-tuning language models on agentic coding patterns found in the conversation logs
• Analyzing tool-use efficiency in Claude Code and Gemini CLI sessions
• Developing benchmarks for coding assistant response accuracy based on real-world conversations

Strengths

• 73 distinct conversation sessions
• MIT licensed for permissive reuse
• Captures modern agentic tool-use patterns from CLI-based assistants

Limitations

• Small sample size of under 100 records
• Potential for inconsistent formatting across different CLI tool exports
• Lack of detailed column descriptions

Provenance

Source

zhiyaowang via DataClaw

Collection Method

Exported from coding assistant CLI tools

Time Range

2026

Freshness

Last updated February 2026.

Geography

United States