Agent Session Manager — Data skill for Claude Code
A modern, fast desktop app for exploring everything Claude Code stores on your machine — sessions, memory, subagents, scratchpads, tasks, analytics and live state — grouped by project.
How to install Agent Session Manager
This entry records only its repository, not the path inside it, so there is no
exact command to give. Open devincii-io/agent-session-manager and copy the folder into
~/.claude/skills/, or the file into ~/.claude/agents/.
What Agent Session Manager does
A modern, fast desktop app for exploring everything Claude Code stores on your machine — sessions, memory, subagents, scratchpads, tasks, analytics and live state — grouped by project. PySide6 + QtWebEngine. Unofficial.
Alternatives in Data
- Qsv — Blazing-fast Data-Wrangling toolkit 3554 17 1 3.6k ★
- Claude Deep Research Skill — Enterprise-grade deep research skill for Claude Code with 8-phase pipeline, source credibility scoring, and au 1k ★
- Bug Bounty Master Workflow — Full pipeline: Recon -> Learn -> Hunt -> Validate -> Report 924 ★
README
Agent Session Manager
A fast desktop workbench for the Claude Code and Codex sessions already sitting on your disk.
**Unofficial community tool.** Not affiliated with or endorsed by Anthropic or OpenAI. Claude is a trademark of Anthropic, PBC. Codex and OpenAI are trademarks of OpenAI. The app indexes local agent session data and nothing else.
What it answers
Every Claude Code and Codex session leaves a `.jsonl` file behind. Nobody reads those. This app turns them into the four questions you actually have.
**How much did this cost, and how much of my time went into it?** The overview opens on spend and hours for the period you pick, each with its trend against the period before, a spend-per-day chart stacked by model or by project, a project table you can sort, and the weekday-by-hour grid of when you really work. Cache savings get their own number, because caching is where the money goes or does not.

**What happened in this session, and what did it mean?** A session opens on a summary: spend, active time versus wall clock, how full the context got, how reliably the tools ran. Under it, findings in plain sentences. Which kind of work ate the hours. Where the agent was idle for three hours. That the context w
Related Skills
BnhassinAgent Database
Persistent Context Across Sessions for Every Agent – Captures everything your agent does during sessions, comp
Agent Telemetry
Live local dashboard for AI coding-tool usage — Claude Code, Claude Desktop, Codex, GitHub Copilot, Cursor, op
Coding Agent Usage
Live local dashboard for AI coding-tool usage — Claude Code, Claude Desktop, Codex, GitHub Copilot, Cursor, op
Verge
Ambient desktop overlay that tracks AI coding agents (Claude Code, Codex/ChatGPT, and more) — real activity, u
Token Lens
Local-only dashboard for AI coding sessions. Track turns, tokens, and cost across Cursor, Claude Code, and Ope
Claude Agent Bus
Claude Code plugin: parallel sessions on one machine see each other, take turns on shared services (dev server
Related Agents
Senior Desktop Engineer
Senior desktop engineer specializing in cross-platform desktop applications, with deep PySide6 (Qt for Python)
Privacy Policy
Last updated: 2026-02-28 nWave is an open-source AI workflow framework that runs inside Claude Code and stores
Broca Memory For Claude Desktop
Get persistent memory in Claude Desktop in under 5 minutes. This package contains everything you need to add f