A native macOS terminal built on libghostty, with vertical tabs, agent-aware notifications and a scriptable CLI for running AI coding agents in parallel.
kenn-io/agentsview
Local-first tool to browse, search, and track token costs across AI coding agents such as Claude Code and Codex.
- Stars
- 6,216
- Forks
- 664
- Open issues
- 112
- Last commit
- 7 hours ago
What is it?
agentsview indexes session history from more than 20 AI coding agents, including Claude Code, Codex, Cursor, Gemini CLI, and OpenCode, into a local SQLite database. It serves a web UI and an HTTP API, and provides a CLI for listing sessions and printing usage reports. The session archive stays on the user's machine unless a sharing feature is turned on.
The tool focuses on three jobs: full-text and optional semantic search across conversation content, token usage and cost tracking with prompt-cache-aware pricing, and analytics such as activity heatmaps, tool usage, and per-project breakdowns. A background daemon handles sync and file watching, and the same archive can be mirrored to PostgreSQL, DuckDB, or ClickHouse.
Support for each agent depends on where that agent stores its sessions, and some agents, such as Aider and Cline, have limits noted in the project documentation. Docker images and desktop builds for macOS and Windows are also published.
Key features
- 01Discovers and syncs sessions from more than 20 coding agents into a local SQLite archive.
- 02Full-text search across messages, with optional semantic search over embeddings.
- 03Token usage and cost reports with prompt-cache-aware pricing and per-model breakdowns.
- 04Analytics dashboard with activity heatmaps, tool usage, and project breakdowns.
- 05Command-line tools for listing sessions, printing usage, and generating stats in JSON.
- 06Web UI with live updates, keyboard shortcuts, and session export as HTML.
Use cases
- Checking how much money and how many tokens you spend across several AI coding agents.
- Searching old agent conversations to find a past decision or fix.
- Reviewing which tools and projects agents spend the most time on.
- Feeding session statistics into scripts or dashboards via the JSON output and API.
Quick start
- 1
Install the CLI with the macOS/Linux install script from the README
- 2
Install the desktop app via Homebrew
brew install --cask agentsview - 3
Start the server in the foreground
agentsview serve - 4
Print the daily cost summary
agentsview usage daily
Things to know
- Install scripts are piped from the web and require trusting a remote installer.
- Requires agents' session data on the local machine; containerized use needs explicit directory mounts.
- Remote access through port forwarding or proxies needs extra configuration.
Topics
FAQ
- What is agentsview?
- Local-first tool to browse, search, and track token costs across AI coding agents such as Claude Code and Codex. agentsview indexes session history from more than 20 AI coding agents, including Claude Code, Codex, Cursor, Gemini CLI, and OpenCode, into a local SQLite database. It serves a web UI and an HTTP API, and provides a CLI for listing sessions and printing usage reports. The session archive stays on the user's machine unless a sharing feature is turned on.
- Is agentsview free to use? What license does it have?
- agentsview is open source under the MIT license. Check the license terms before using it in commercial products.
- Is agentsview still maintained?
- Actively maintained. The last commit to agentsview was on Oct 21, 2026.
- What can I use agentsview for?
- Checking how much money and how many tokens you spend across several AI coding agents. Searching old agent conversations to find a past decision or fix. Reviewing which tools and projects agents spend the most time on. Feeding session statistics into scripts or dashboards via the JSON output and API.
- How do I install agentsview?
- Install the CLI with the macOS/Linux install script from the README Install the desktop app via Homebrew: brew install --cask agentsview Start the server in the foreground: agentsview serve Print the daily cost summary: agentsview usage daily
- Where can I get agentsview?
- The source code and installation instructions are on GitHub: https://github.com/kenn-io/agentsview
Summary generated by AI from the project's README. GitHub stats refresh daily.
Related projects
Desktop workspace that runs many coding agents in parallel, each on its own git branch and worktree, with a live Kanban of PRs, CI and reviews.
Local CLI and desktop app that breaks down your AI coding token spend and cost by tool, model, project and task.
An Agent Skill that lets coding agents write and run custom Playwright browser automation scripts for tests and multi-step web tasks.
A terminal UI for viewing, checking, installing and updating agent skills across Claude Code, OpenCode, Codex, Cursor and other coding agents.
Persistent memory for coding agents that captures session activity, compresses it with AI, and injects relevant context into future sessions.