Reading
agwatch is a terminal analysis CLI tool designed specifically for AI programming workflows. It reads local AI assistant usage data, standardizes it into a unified event model, and presents it via an interactive dashboard or structured summary, helping developers accurately track the costs and efficiency of AI programming.
Section 01
agwatch is a terminal analysis CLI tool designed specifically for AI programming workflows. It reads local AI assistant usage data, standardizes it into a unified event model, and presents it via an interactive dashboard or structured summary, helping developers accurately track the costs and efficiency of AI programming.
Section 02
With the popularity of AI programming assistants like OpenCode and Claude Code, developers are increasingly relying on these tools to boost coding efficiency. However, this dependence also brings a practical problem: how to clearly understand the input and output of their AI programming efforts?
How much does each call cost? Which project consumes the most tokens? Which tools are used frequently? The answers to these questions are often scattered across local data files of various AI assistants, making it difficult to aggregate and analyze. agwatch was born to address this—it integrates these scattered data sources and provides a unified view to gain insight into the entire AI programming workflow.
Section 03
agwatch is designed to answer the practical questions that developers care about most:
With this data, developers can make more informed decisions—such as optimizing prompts to reduce token consumption or identifying which projects benefit most from AI assistance.
Section 04
The core innovation of agwatch lies in its definition of a unified usage event model. Different AI assistants (like OpenCode and Claude Code) use different local storage formats (SQLite, JSON, JSONL). agwatch reads these data through an adapter layer and converts them into a standardized internal representation.
This design offers several benefits:
Currently, agwatch supports:
Section 05
agwatch's main feature is its interactive terminal dashboard. Upon launch, users will see a multi-panel interface that displays various metrics of AI programming activities in real time:
Section 06
Section 07
Displays cost, tokens, and session count grouped by project, helping identify which codebases are the most active.
Section 08
Displays cost, tokens, and call count grouped by model, revealing which LLMs are your mainstays.
Keep going with more reads from the same topic.
Splinter is a minimalist, high-performance key-value (KV) and vector storage system enabling zero-latency inter-process communication via shared memory and atomic operations. With only 766 lines of core code, it supports millions of operations per second and 768-dimensional vector storage, offering a new architectural approach for local LLM inference and data-intensive applications.
Folkering OS is the world's first AI-native bare-metal operating system, entirely written in Rust no_std without relying on Linux, POSIX, or libc. It can generate commands from scratch, compile them into WASM, and run them in 10 seconds, achieving true self-evolution.
This project explores how to use Large Language Models (LLMs) to detect logical vulnerabilities in Ethereum smart contracts, especially business logic flaws that are difficult for traditional static analysis tools to capture, such as economic manipulation and execution order errors.
Cross-generational hardware testing based on NVIDIA A10G and H100 GPUs, comparing and analyzing the throughput, latency, and concurrent scalability of three mainstream LLM inference frameworks. SGLang achieves a 3.4x performance advantage over vLLM on H100, while Ollama faces architectural bottlenecks in high-concurrency scenarios.