Claw Diary: Visualize AI Agent Activity for Enhanced Insights

Claw Diary offers a comprehensive solution for monitoring and visualizing AI agent activity. By capturing every tool call made by an AI agent, the platform provides valuable insights into its behavior, performance, and cost. The core functionality revolves around transforming raw activity data into various formats, including daily narrative summaries, visual timelines, and cost analytics. The daily diary provides a markdown summary of the agent's activities, including session breakdowns, tool usage, and key insights. The visual timeline offers an interactive HTML page with search, filtering, and animated replay capabilities, allowing users to easily explore the agent's actions over time. The cost analytics feature provides 30-day trends, per-model breakdown, and failure rate tracking, enabling users to optimize their AI agent deployments for cost efficiency. Claw Diary's local-first design ensures data privacy and eliminates the need for external API calls. All data is stored locally in JSONL files, providing users with full control over their data. The platform's template-based summaries eliminate the need for LLM calls, further reducing costs and ensuring data privacy. The installation and configuration process involves using command-line tools and manually editing configuration files. Claw Diary stores all data as daily JSONL files under `~/.claw-diary/events/`. No database required. Template-based summaries — No LLM calls for narrative.