chat_logger

What This Skill Does

Chat Logger is a specialized OpenClaw agent skill designed for robust, high-reliability message tracking across communication platforms like Feishu and DingTalk. It functions as a dual-layer logging system that combines automated message hooking with mandatory explicit calling to ensure zero data loss. Unlike generic logging tools, this skill implements a strict command-trigger architecture, preventing the main OpenClaw agent from attempting to process specific analytical queries (like "今日简报") by mandating an immediate handover to the chat_logger engine. This decoupling ensures that data retrieval remains consistent, clean, and isolated from the agent's generative reasoning loops.

Installation

To install this skill, run the following command in your terminal: clawhub install openclaw/skills/skills/bjmfjoy/chat-logger Ensure your agent environment is configured with write permissions for the local file system, as the skill periodically writes chat logs to local storage. After installation, verify the setup by running the built-in health check function, which validates the file structure and write access.

Use Cases

1. Corporate Knowledge Management: Automatically log internal team communications for historical reference without manual intervention.
2. Daily Performance Tracking: Use the "今日简报" command to get a summarized overview of today's interactions, helping team leads maintain situational awareness.
3. Audit and Compliance: Maintain a secure, structured record of private direct messages, ensuring that important business decisions made via chat are captured and indexed.
4. Personal Productivity: Use the "我的记录" feature to quickly retrieve your own interaction history for project reflection or follow-up tasks.

Example Prompts

1. "chatlog汇总"
2. "今日简报"
3. "我的记录"

Tips & Limitations

• Strict Adherence: Developers must strictly follow the provided AGENTS.md guidelines. If the main agent processes queries like "今日简报" instead of delegating, the logs may conflict with the agent's logic. Always return immediately after calling handle_chat_query().
• File System: Ensure the host server has adequate disk space for persistent logging. The skill creates multiple files based on the channel and user context.
• Security: Because this skill performs file-read and file-write operations, ensure that your OpenClaw environment is running in a secure, restricted directory to avoid unauthorized log access.