Agent infrastructure
tokentap
jmuncor/tokentap
A local proxy and live terminal dashboard that intercepts LLM API traffic from Claude Code, Codex and similar CLIs to track token usage, context-window consumption, costs and archived prompts.
Topics
Related projects
claude-tap3.3kA local proxy and trace viewer for AI coding agents that captures and compares real API traffic, prompts, tool calls, and token usage from Claude Code, Codex, and others.
crit1.3kCrit is a local-first review tool for AI coding agents. It renders plans, diffs, and running apps in the browser so you can leave line-level comments that are sent back to the agent.
cc-lens601A local analytics dashboard for Claude Code that visualizes sessions, tokens, costs, and tool usage from your own machine.
any-llm2.2kA Python SDK that lets developers call any LLM provider through one unified, framework-agnostic completion interface.
easy-agent1kEasy Agent is a production-ready terminal coding agent with permission rules, OS sandboxing, MCP, skills, sub-agents and multi-provider models, designed to be readable and extensible.
claude-code-hooks576A marketplace of installable hooks for Claude Code that enforce runtime safety, protect secrets and Git branches, format code, log sessions and surface cost or context insights around the coding agent.