AgentsView | AgentsView<br>Skip to content

AgentsView<br>A local-first desktop and web app for browsing, searching, and analyzing your past AI coding sessions. See what your agents actually did across every project.

Get Started View on GitHub

PiebaldQClawQwen CodeWorkBuddyAntigravity

Quick Start

Download the desktop app (recommended):

Download the latest .dmg (macOS), .exe (Windows), or<br>.AppImage (Linux) from<br>GitHub Releases or via homebrew: brew install --cask agentsview.<br>The desktop app is fully bundled and includes built-in<br>auto-update.

Install via pip — or run instantly with uvx:

Terminal windowpip install agentsview # install permanently

uvx agentsview # or run without installing

Install via shell script:

Terminal windowcurl -fsSL https://agentsview.io/install.sh | bash

Windows (PowerShell):

Terminal windowpowershell -ExecutionPolicy ByPass -c "irm https://agentsview.io/install.ps1 | iex"

Terminal windowagentsview serve # Start server

agentsview serve --port 9090 # Custom port

agentsview serve --no-browser # Disable browser auto-open

Note<br>The desktop app and CLI share the same data directory<br>(~/.agentsview/), so you can use one or both — they are<br>fully complementary.

Fast Token Usage & Cost Reports

If you’ve been reaching for<br>ccusage to see how<br>much you spent on Claude Code yesterday, try<br>agentsview usage instead. It reads from the<br>same pre-indexed SQLite database that powers the rest of<br>AgentsView, so reports come back in well under a second even on<br>large histories. As of 0.32.0 it reports on Claude Code ,<br>Codex , Copilot CLI , OpenCode , Pi , Gemini ,<br>and Hermes sessions, with more agents landing as their<br>parsers start recording token usage.

Terminal windowagentsview usage daily # last 30 days, terminal table

agentsview usage daily --all # full history, JSON-friendly

agentsview usage statusline # $9.61 today

On a 22,000-session local database, agentsview usage daily<br>runs 80–220× faster than npx ccusage@latest daily (see<br>benchmarks). On<br>smaller databases the absolute gap is smaller, but reports<br>still come back sub-second. See<br>Token Usage & Costs for the full write-up.

What It Does

AgentsView reads the session files that your<br>AI coding agents leave on<br>your machine and gives you a local-first desktop and web app to<br>work with them. By default everything stays on your machine.<br>Optionally, PostgreSQL sync can push session data<br>to a shared database for team or multi-machine setups.

AI-Powered Insights<br>Generate summaries and analysis of your coding sessions<br>using Claude, Codex, Copilot, or Gemini. Get daily<br>activity digests, multi-day analyses, and<br>recommendations — scoped by project or across everything.

Browse Sessions<br>Scroll through every session across all your projects.<br>See the full conversation: user prompts, assistant<br>responses, thinking blocks, and tool calls. Filter by<br>project, agent, date, or message count.

Search Everything<br>Full-text search across all message content. Find that<br>one conversation where you discussed a specific function,<br>error message, or design decision — even months later.

Analyze Your Usage<br>Activity heatmaps, tool usage breakdowns, velocity<br>metrics, session-health analytics, per-project stats,<br>and session distribution charts. Understand how you use<br>agents over time.

Token Usage & Costs<br>A sub-second agentsview usage CLI for<br>daily spend reports and a today’s-cost status line. A<br>ccusage alternative for Claude Code, Codex, Copilot CLI,<br>OpenCode, Pi, Gemini, and Hermes sessions that runs 80–220×<br>faster on large session histories.

Live Sync<br>Watches your session directories for changes and<br>streams new messages in real time. Start a coding<br>session in one window, watch it appear in AgentsView<br>in another.

Multi-Agent Support<br>Works with 27 AI coding agents<br>including Claude Code, Codex, Copilot, Cursor, Gemini,<br>and OpenHands. Auto-discovers session directories so<br>there’s nothing to configure.

Import Chat History<br>Import your Claude.ai and ChatGPT<br>conversations — including images. Upload a zip export<br>and browse everything in one place alongside your<br>agent coding sessions.

Runs Locally<br>SQLite database, embedded web frontend, no cloud<br>services, no accounts. Install the desktop app or<br>a single binary and run it.

How It Works

AgentsView watches your agent session directories for changes,<br>parses JSONL files from each agent format, and stores structured<br>data in SQLite with full-text search indexes. The embedded web<br>frontend provides browsing, search, and analytics over the<br>REST API.