
agent-standup
Your Claude Code, Codex and Mistral Vibe sessions as a live, animated team call: who is working, who waits on whom, who needs you. VS Code extension, terminal c
Sign in to rate this tool.
Overview

Agent Standup visualizes the activity of Claude Code, Codex and Mistral Vibe agents as a live, animated team call. The tool reads the session files that your agents already write and presents them as a meeting you can watch in real time or replay later. It runs locally, is free and open source, and can be started from VS Code, Cursor, any terminal, or a desktop application.
The core experience is a call‑style interface where each agent occupies a desk, speaks with a moving mouth, and shows a single status: working, waiting on someone, needs you, or done. When an agent runs a command you see the command line, its output, and any file changes as a diff. Failed commands are highlighted, making the first error easy to spot. You can scrub through a past session, jump to the longest wait or a loop, and at the end a recap lists agents, commands, errors, token usage and changed files. A board view aggregates every session across projects, and you can export the call as a video.
Agent Standup is distributed as a VS Code extension (v1.42.0 or later), an Open VSX extension for Cursor, Windsurf or VSCodium, a terminal command (`npx @double_geste/agent-standup @latest --open`), and a native desktop app for macOS, Windows and Linux. The VS Code panel adds live counts in the status bar that turn red when a session waits for your input. The desktop app shows a menu‑bar indicator and a notification for the same condition. All variants run a small local server; no data is sent elsewhere unless you choose to export a video.
The interface supports natural‑sounding voices in ten languages, and you can add custom character images by dropping them into a folder. An optional Halloween pack adds eight hand‑drawn characters for a small fee that supports the artist. The tool works with Claude Code, Codex, Mistral Vibe and the Cowork sessions of the Claude desktop app, requiring that you have run at least one session on the machine before launching Agent Standup.
Key features
- Reads local Claude Code, Codex and Mistral Vibe session files and visualizes them as a live call
- Shows command output, file diffs and error markers for each agent
- Replay mode with scrubbing, jump‑to‑error and end‑of‑session recap
Details
- Pricing
- Free
- Category
- Code
Suggest an edit
Sign in to suggest an edit.
Pros and cons
Pros
- Runs entirely locally, no token usage or external API calls
- Provides a visual overview that reduces the need to read raw logs
- Multiple entry points let you use the tool in the editor you prefer or without an editor
- Open source code is available for inspection or modification
Cons
- Requires Node 18 or later for terminal and desktop usage
- Desktop app is not signed by Apple, requiring manual security override on macOS
- No pricing information is published because the tool is free, but optional packs cost extra.

