Show HN: Lookspan – local-first observability for AI agents (npx lookspan) (github.com)

Lookspan has launched a local-first observability dashboard tailored for AI agents, allowing developers to track every operation (or "span") emitted by their agents without relying on cloud infrastructure. Unlike traditional observability tools that require cloud accounts and data sharing, Lookspan operates entirely locally, ensuring data privacy and eliminating external costs. By simply running `npx lookspan`, users can instantly visualize performance metrics in a web browser, with support for various frameworks like MCP, LangGraph, and OpenTelemetry. This tool is significant for the AI/ML community as it enables real-time monitoring and debugging of AI agents with minimal setup. Users can easily ingest span data through HTTP requests, track costs associated with token usage, and receive alerts for unexpected behavior. Additional features include replaying interactions, scoring outputs with LLM judges, and managing datasets for batch evaluations. Lookspan emphasizes security and user-friendliness, making it an attractive option for developers looking to enhance observability and control over their AI systems, especially in environments where privacy and local data management are critical.