tzafrir/whats-agent-doing
What's Agent Doing: a Claude Code plugin with a little box above the prompt that always says what Claude is doing right now - tzafrir/whats-agent-doing
1 stars · 0 forks
README
What's Agent Doing See what Claude Code is doing at every step, in plain English, live. A box above the prompt names the step Claude is on and how long it has been on it. Click ▸ to see every step it has taken. Why You ask Claude Code to fix a bug, and for the next few minutes you see a spinner and long commands scrolling past. You can't easily tell:
whether it's stuck or just slow, whether it's waiting for you, what that command is for, and whether it worked.
A bug fix, step by step
The cart total is wrong when a coupon is applied. Fix it.
To track this down, Claude runs commands you'd have to stop and decode. The box says what each one is for, and how long it has been running:
Claude runs The box says
npx tsx -e "import {cartTotal} from './src/cart'; console.log(cartTotal({…coupon:{amount:10},taxRate:0.08}))" Reproduce the total with a $10 coupon
git bisect start HEAD v2.3.0 && git bisect run npx vitest run src/cart -t "fixed coupon" Find the commit that broke fixed-amount coupons · 1m 48s
git bisect reset && npx vitest run src/cart Waiting for your approval: Reset the bisect and rerun the cart tests
The bisect runs for almost two minutes. The box shows it is bisecting, not stuck. When Claude needs you, the box turns yellow, so a wait for approval never looks like a hang. Click ▸ for every step and how it went: ╭─────────────────────────────────────────────────────────────╮ │ ▾ ○ Claude: Idle · last turn done in 2m 40s, 5 actions │ │ │ │ › The cart total is wrong when a coupon is applied. Fix it. │ │ ✓ Reading cart.ts 0.6s │ │ ✓ Reproduce the total with a $10 coupon 2.1s │ │ ✓ Find the commit that broke fixed-amount coupons 1m 48s │ │ ✓ Editing cart.ts 0.6s │ │ ✓ Reset the bisect and rerun the cart tests 7.4s │ │ ✎ Wrote the reply (47 words) 5.8s │ │ ✓ Done in 2m 40s, 5 actions 2m 40s │ ╰─────────────────────────────────────────────────────────────╯
You wonder Without With the box
Is it stuck? A spinner and a long command line The step in plain English, and how long it has been on it
Is it waiting for me? A dialog you see only if you're looking A yellow box: Waiting for your approval: …
What is that command for? Decode the flags yourself The command's own description: Find the commit that broke fixed-amount coupons
What happened while I was away? Scroll back and piece it together ▸ history with ✓ done, ✗ failed, ⊘ denied, ■ interrupted, and durations
Background agents get a row each, and keep it after your turn ends, while they're still working: ╭───────────────────────────────────────────────────────────────────╮ │ ▸ ● Claude: 2 agents working · 34s │ │ ◆ Find callers of applyCoupon › Searching for "applyCoupon(" 34s │ │ ◆ Check the coupon tests › Reading coupon.test.ts 33s │ ╰───────────────────────────────────────────────────────────────────╯
The plugin is free and MIT-licensed. It only displays: no network, no files, no telemetry. It is one of the first mods, built on Claude Code's new function hooks. Install In Claude Code: /plugin marketplace add tzafrir/whats-agent-doing /plugin install whats-agent-doing@tzafrir
Or from your shell: claude plugin marketplace add tzafrir/whats-agent-doing claude plugin install whats-agent-doing@tzafrir
Then start a new session (or run /reload-plugins). Update later with claude plugin marketplace update tzafrir. Requirements: Claude Code in a terminal. The plugin is a mod, written with Claude Code's function hooks, which are early access: they load only where function hooks are enabled, and their API may change between releases. The box is drawn in the band above the prompt, which only the terminal has; hooks are not loaded in Claude chat or Cowork. What it shows
State Example
Reading the prompt Reading your prompt
Thinking: the thought's latest sentence where the model streams it, else the last action with the dot turned magenta Thinking: …the hook needs to await next
Writing a tool call, once its input says what it does (a file's path, a command's description) Writing register.tsx · 2.3 KB
Running a tool Reading app.ts, Find the commit that broke fixed-amount coupons (a command's own description), Fetching github.com
A subagent at work Running an agent: find the bug › Reading app.ts
Background agents, during a turn or after it 3 agents working, with a row per agent: ◆ Draft the plan › Searching the web 1m 12s
Waiting on you Waiting for your approval: Reset the bisect and rerun the cart tests, Waiting for your answer
Reading tool results Reviewing the results of 3 actions
Writing the reply Writing the reply · 120 words
Compacting Compacting the conversation
Between turns Idle · last turn done in 42s, 6 actions (or interrupted, stopped on an error, declined)
Parallel calls show the newest with (+N more), and a timer counts how long the current state has lasted. Click the triangle (or focus the band with ctrl+x tab and press Enter) to expand the history: each prompt, stretch of thinking, tool call (✓ done, ✗ failed, ⊘ denied, ■ interrupted), reply and turn ending, with durations. The box steps aside while a survey holds the band; collapse the whole band with ctrl+x ctrl+a. Data and privacy
Stays in the session. The plugin keeps its headline and history in the session's own state ($.state) and nothing else: no files, no network, no processes, no storage across sessions. claude plugin validate lists every call it makes: $.clock, $.state and $.ui. Only displays, never changes what Claude does. Every hook is display only and changes nothing: session.start, turn.start, turn.step (the model's stream), tool.call, classic.PermissionRequest, classic.SubagentStop, classic.PreCompact, classic.PostCompact and turn.complete each read their event and pass it on as it came. Settings, instructions, tool descriptions, other hooks' results, tool calls and their results, and the model's stream all reach Claude exactly as they would without the plugin. Only displays permission prompts, never answers them. The plugin hooks classic.PermissionRequest, which runs each time a tool call, Claude's or a subagent's, needs your approval. The hook is display only and decides nothing: it never allows, denies or changes the request, adds no permission rules, and passes the request on as it came. Its one job is to show Waiting for your approval: <tool> in the box until you answer. Shows what Claude is working on. The box and its history display your prompts, file paths, URLs, the descriptions of commands (or the command itself when it has none), MCP tool names and, where the model streams it, a line of its thinking. Anyone who can see your screen, or a screen share, can read them. Labels are cleaned of control and invisible characters before they are drawn.
Read the full privacy policy. Support Report a problem or ask for a feature in GitHub Issues. Develop Load the plugin from a clone; the session reloads it as you edit: git clone https://github.com/tzafrir/whats-agent-doing claude --plugin-dir whats-agent-doing
claude plugin validate . claude plugin test .
Loading the plugin writes the API's types to .claude-plugin/types/, which tsconfig.json extends; then npx -p typescript tsc -p . typechecks it.
File What it does
hooks/register.tsx The hooks: the model's stream (turn.step), tools (tool.call), approvals (classic.PermissionRequest), subagents (agent.spawn, their calls and stops), compaction, turns, and the box's drawing (ui.render on AbovePrompt)
hooks/activity-of.ts Turns a tool call, whole or still streaming, into its short label
hooks/text.ts Printable labels, thought snippets, word counts, sizes, durations
types/index.d.ts The $.state contract: the headline, the agents at work, the history, the toggle
tests/register.test.tsx Idle, a whole turn state by state, the history toggle, thinking and commands without repeats, hostile labels, an approval wait, background agents after the turn
.claude-plugin/marketplace.json Makes this repository its own one-plugin marketplace
.claude-plugin/icon.png The listing icon: the box and its status dot above the prompt
License MIT