About clioilc

Observability for AI coding agents

Our Mission

AI coding agents now write, run and fix a large share of the code we ship. When an agent works for an hour, the final diff tells you little about what it actually did, where it got stuck, or what it cost. clioilc exists to make that work visible.

What We Do

clioilc reads the session logs your agent already keeps on your machine and turns them into a clear record: every prompt, reply and tool call, which ones failed and why, the tokens used and what they would cost on the API, per project, model and day. You can read it in the terminal, or sync it to your dashboard at dash.clioilc.com.

We start with Claude Code. Support for Gemini CLI and Codex is on the way, and readers for other agents are plugins anyone can write.

Why It Matters

  • Transparency — see what your agent did, step by step, not just the final diff
  • Quality — spot failing tool calls and API errors as they happen
  • Cost — know which projects and models your tokens go to
  • Privacy — your prompts and replies stay yours, encrypted end-to-end if you choose

Open Source

The clioilc CLI that runs on your machine is open source, so you can check exactly what it reads and what it sends: github.com/PUYUP/clioilc-tracker.

Get in Touch

Have questions or want to learn more? Read the docs or contact us.