Agent Standup

See who is waiting on whom, and who is blocked — across all your Claude Code sessions — at zero token cost. Agent Standup turns the session files Claude Code already writes into a live team call (and a board of every session): it only reads local files and never calls Claude unless you turn on French translation or the jokes.

Your Claude Code session as a live team call. The lead gives orders, sub-agents join the call, work at their desks, share their screen and report back — with voices, expressions and a view of the real work as it happens.

→ Website & download: https://atoll92.github.io/agent-standup/

A fictional team fixing a slow login page — the built-in demo (⚙ → Source → Demo). Watch the full 1-minute video.

Unofficial community project, not affiliated with or endorsed by Anthropic. "Claude" is a trademark of Anthropic.

What you get

• Token usage per agent — a ⚡ counter in the header and on each tile shows how many tokens each agent used (input + output + cache writes; cache reads shown separately in the tooltip), counted once per message as recorded by Claude Code. Handy to see which sub-agent is the expensive one. It is read from the session files, not an estimate of your bill.
• Stuck or quiet agents — in the live call, a tile pulses amber with ⏳ when a command has been running for over a minute, shows 💤 when a working agent has gone quiet for 3 minutes, and a red ✗ counter for failed tool calls. The browser tab title shows how many agents are working.
• Built-in demo — no session yet? ⚙ → Source → Demo plays a fictional team (no data of yours).
• Live call — follows your newest Claude Code session in real time. Each agent is a small character at its desk; the one who speaks gets the spotlight and keeps it until someone else talks. Agents that talk back-to-back talk over each other, like a real meeting.
• Real voices, lip-sync — lines are spoken aloud; mouths follow the actual volume of the voice.
• Live screen sharing with the real output — when an agent works, its screen fills the stage (presenter's camera in the corner, "sharing" badge): a terminal with the command history and the actual output of each command, search results and tool calls; or an editor with the file it reads (real content and line numbers) or the exact diff it applies. See What you see for each agent below.
• Agents come and go — a sub-agent joins when it is given a task and leaves the call 2 minutes (playback time) after its report, with a dissolve animation. It comes back if it gets a new task.
• Sub-agent tree — the full hierarchy (you → Claude → sub-agents → their sub-agents) with status, tool counts and the current action.
• Click any agent — its prompt, tool usage and reports.
• Replay — pick any past session, scrub through it, see a timeline of who worked when, and get a recap (agents, tool calls, commands, files changed, busiest agents).
• Video export — records the call as a 1280×720 .webm with voices and subtitles.
• French or English — agent text is translated on the fly (see below).
• Office music — optional live web radio (FIP and its themed stations) that fades down smoothly while an agent talks and fades back up when the room goes quiet.
• Office ambience, optional (off by default) and fully synthesised: keyboards that type when agents run commands, clicks when they read, a coffee machine, a phone buzzing, a far-away door… (⚙ → Office ambience). Plus small reactions and jokes between agents ("episode mode").

What you see for each agent

On its tile

• Activity tag (top right) — what the agent is doing right now: Reading App.jsx, Editing foo.ts, Writing …, Running npm test (or the command's own description), Searching “pattern”, Listing **/*.js, Fetching host, Updating todo list; then ✓ done when it reports.
• Ticker (bottom) — the current tool call and target, with a spinner and a running counter while it executes, then ✓ / ✗ and its real duration.
• Status border: orange while working, spotlight while speaking, the tile fades and dissolves 2 minutes after its report.

On the shared screen (appears when an agent uses a tool, stays while a command is still running, then ~4 s after the last action)

• Terminal — ❯ command (with its description), then its real output (first lines, errors in red, “… +N lines”), ● Grep(…) / ● Glob(…) with matches, a spinner for what is still running. Footer: ✓ Bash · 1.2 s or ⠋ Bash · running 3.4 s.
• Editor — for Read: the file with its real content and line numbers (read-only); for Edit: the removed/added lines; for Write: the new content, with the tool's result line.
• Task list — if the agent keeps a todo list, a side panel shows it live (☐ pending, ◐ in progress, ☑ done).
• The presenter's camera lip-syncs when they speak; the subtitle bar shows what they say.

Click an agent — its prompt, tool counts and a Recent activity list (last 30 calls with status, duration, and expandable command + output).

Outputs are trimmed (about 1–2 KB each) to keep the call fast. Results of internal calls (sub-agent launches, messages) are not shown.

Screenshots

Real commands, real output	The exact change	Who asked whom

Task list of an agent	End-of-episode recap	In French

Personality, without noise

Sub-agents answer when they get an order ("Running the tests, as usual.", "Grepping like there's no tomorrow."), say goodbye when they leave, and now and then have a little thought; a failing tool call makes the tile shake; confetti when the lead finishes a long job. It is all local (no API), absent from the transcript, and can be turned off in ⚙ → Little lines & animations.

Glance without the panel (VS Code)

The status bar shows live counts for your workspace (needs you, working, your turn) and turns red when an agent may be waiting for you; hover for the sessions, click to open the call on the most urgent one (agentStandup.statusBar).

Context gauge, highlights, loop watch

The header shows how full the lead's context is (ctx 62%, orange then red, with a warning before auto-compaction); tiles and list rows show it from 50%. In Replay, the Highlights bar jumps to the first error, the longest wait, compactions and possible loops. A 🔁 badge marks an agent repeating the exact same call or failing repeatedly. The context window (200k or 1M) is inferred, because transcripts do not state it.

Languages

Pick one in ⚙ → Language: English, Français, 繁體中文, 简体中文, Español, Deutsch, Português (Brasil), 日本語. The interface (menus, chips, tooltips, recap) is translated; VS Code's command titles, setting descriptions and status bar follow VS Code's own display language. English and French are written by hand; the other languages are AI-drafted and not yet reviewed by native speakers, so corrections are very welcome: each language is one small JSON file (public/locales/<code>.json), see public/locales/README.md to fix or add one (npm run i18n checks the files). The words the agents say stay in the language of your session (French translation of their lines is optional); the little banter lines and the demo exist in English and French.

Your own characters

Replace the drawn characters with your own pixel art. Drop PNG (or GIF/WebP) files, 300 KB at most each, into ~/.agent-standup/characters (⚙ → Characters shows the folder, and in VS Code Open folder opens it; the agentStandup.charactersPath setting changes it):

• lead.png (or claude.png) is the lead, you.png is you, any other image becomes a sub-agent (one per agent, repeated if you have fewer images than agents);
• name-talk.png next to name.png is an optional open-mouth frame that shows while the character speaks;
• 32×32 or 64×64 px, transparent background, facing front, feet at the bottom looks best. The art is shown crisp (no smoothing) on top of the usual scene and desk, and every status, hand and badge still works. Press Reload in the menu to see changes. A starter pack is in examples/characters.

Agents list

On wide windows (900 px+), a fixed list sits left of the call: every agent with its role and one status line, most urgent first (needs you, stalled, working, waiting, done) and the lead pinned on top. It stays readable with dozens of agents, which a grid of tiles does not. Click a name to put that agent in the spotlight (kept for 25 s even when someone else speaks), double-click for details. Toggle it in ⚙ → Agents list.

All your sessions (Sessions view)

The Sessions tab lists every session active in the last 12 hours, across all projects, with one state each: working, needs you?, waiting on sub-agents, your turn, or idle (and how many sub-agents are active). It refreshes every few seconds and reads only the end of each transcript, so thirty parallel sessions stay cheap. Click a card to follow that session in the call.

At a glance

Every tile carries one status: working (with what it is doing), **waiting on another agen