
Every AI coding agent, every token, one local dashboard
Local-first token observability for every AI coding agent: Claude Code, Codex CLI, Gemini CLI, OpenCode and any agent via API. Debug prompts, context and cost per request, project, tool, skill and MCP server, with a REST API, Python/TypeScript SDKs, OpenTelemetry export and webhooks. One command: npx knowyourtokens.
A source that found nothing is a measurement. A source that has not run is a gap. Neither means the launch lacks the thing.