staceb

Claude Session Explorer — Development skill for Claude Code

Development community

Second Brain: a private, local dashboard for your Claude Code history.

How to install Claude Session Explorer

This entry records only its repository, not the path inside it, so there is no exact command to give. Open staceb/claude-session-explorer and copy the folder into ~/.claude/skills/, or the file into ~/.claude/agents/.

What Claude Session Explorer does

Second Brain: a private, local dashboard for your Claude Code history. Mind map, sessions, agents, what's installed, and token usage. Runs on your machine and your own Claude plan.

Alternatives in Development

  • Agentlas Update — Update the installed Agentlas runtime and every host adapter on this machine 1.1k ★
  • Claude Usage — A local dashboard for tracking your Claude Code token usage, costs, and session history 864 ★
  • Dev Health — Check {{DEV_MACHINE}} dev environment health dashboard 57 ★

README

Second Brain (claude-session-explorer)

**A private, local dashboard for everything you've done with Claude Code: what you worked on, what ran, what's installed, and how many tokens it all used.**

Second Brain: hovering the mind map, opening a conversation and a theme

▶ Watch the video (MP4)

Overview, charcoal Overview, bright
Overview dashboard on the charcoal background Overview dashboard on the bright yellow background

Claude Code keeps a transcript of every session in `~/.claude/projects/`. Second Brain reads them, plus Claude Code's own config, and gives you:

  • Overview: tokens this month, sessions open right now, agent runs, what loads before you type, tokens per week, and your heaviest sessions.
  • Map: your conversations, memory notes and automated jobs, grouped into themes (active, simmering, parked) and linked by similarity, plus connections Claude noticed between themes.
  • Sessions: conversations and automated runs past and present, the sessions open right now (with memory use and whether they're working), and your heaviest requests ranked by tokens. Each session has a summary card, its token breakdown, and the full readable transcript.
  • Agents: every agent that has run, which session launched it, how many tokens it used, and who decided to launch it (you, Claude on its own, another agent, or a script). Includes the agents you've defined and which are running now.
  • Installed: skills, connectors and MCP servers, plugins, scheduled jobs, memory and instructions, each with its token load and how often it's been used. You can archive any of them and restore it later.
  • Memory: everything Claude remembers, each note linked to the session that wrote it.
  • Search across everything you and Claude