Can I afford this refactor? Check your plan limits first

ACTVT reads your Claude and Codex plan usage from the provider: how much of the 5-hour session window and the weekly window is used, and when each resets. Agents can ask for it over MCP with the agent_usage_limits tool before starting a long job, and the same usage limits are in ACTVT’s Overview. Plan usage meters are part of the Premium tier.

0:484 chapters

Step by step

  1. 0:00Ask before you start

    A two-hour job is ahead, on a plan with usage limits. Before starting, ask Claude Code whether it fits.

  2. 0:15Claude checks the plan

    Claude asks ACTVT for the plan’s current usage. It fits: 8% of the 5-hour window is used, and 40% of the week. Each window comes with its reset time, read from the provider.

  3. 0:26The same numbers in ACTVT

    The same windows are live in ACTVT’s Overview: Claude’s session and weekly limits with reset countdowns, Codex’s weekly window beside them, and the plan each one is on.

  4. 0:42Spend, honestly labelled

    ACTVT prices spend is at API rates. On a subscription plan that is not what you actually pay, and ACTVT labels it correctly.

More tutorials

All tutorials →
1:24

Every machine you code on

See agent sessions from every machine you code on, over SSH

Start Claude Code on another Mac over SSH, watch the session live from your own, and read the full transcript when it finishes.

3:42

Leftover ports

Find and kill the dev server your AI agent left running

Claude Code starts a dev server, quits, and leaves port 3000 taken. ACTVT spots the orphan and frees the port safely.

2:14

Agent or machine?

Was it the AI agent, or my Mac? Diagnose a slow agent run

A Codex build felt slow. Claude asks ACTVT what the Mac was doing during that session, and gets an answer backed by numbers.

Try it on your own agents

ACTVT is free to start, and reads the session files Claude Code, Codex CLI and Pi already write. Nothing to configure.