What did my AI agent remember from earlier sessions?

ACTVT reads each Claude Code transcript for memory. It shows which memories were listed in the memory index when the session started (Knew), which memory files the agent opened, and which ones it created or updated. The transcript marks each point where a memory was saved or opened, the inspector counts Knew, Opened and Changed, and Show memories lists each memory with what happened to it and the lines a session wrote. ACTVT only reads the transcript. It never changes your memories.

0:343 chapters

Step by step

  1. 0:00Open the session

    Claude skipped the e2e tests here, and nobody asked it to in this session. It says the e2e sandbox only works in CI. So how did it know that?

  2. 0:09What it knew

    It started the session knowing one memory. It shows Knew 1 and Opened 0, so it never even opened the memory file. When I click Show memories, you can see the one-line summary in the index was enough.

  3. 0:25Where it came from

    An earlier session saved that memory, right after its e2e test failed. Show memories on that session shows the lines it wrote.

More tutorials

All tutorials →
1:24

Every machine you code on

See agent sessions from all your machines over SSH

Start Claude Code on another Mac over SSH, and you can follow the session from your own Mac while it runs and read the whole transcript when it’s done.

3:42

Leftover ports

Find and kill the dev server your AI agent left running

Claude Code starts a dev server and then quits, and port 3000 stays taken. ACTVT shows you the server whose session is gone and lets you free the port safely.

0:48

A plan-aware agent

Check how much of your plan is left before a long agent job

Before a two-hour job, Claude Code asks ACTVT how much of your plan is left. You can see the same limits in ACTVT, along with when each one resets.

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.