SLOPSHOPPER

task-inbox

To-do pane: Claude's live plan, your own to-dos, and an inbox of tasks you dispatch to background subagents in one press.

newpaneguardcommandstatustimer
A shopper browsing a rack in a slop shop
Preview · a replayed session in a sandbox
claude · ~/work/app · task-inbox
│ ┃ Tasks ✕ › fix the failing auth test and add an audit log call │ ┃ Claude's plan │ ┃ No plan yet. It fills in as Claude tracks ⏺ Read(src/auth.ts) │ ┃ its steps. ⎿ Read 6 lines │ ┃ ⏺ Update(src/auth.ts) │ ┃ My to-dos (0) ⎿ Added 2 lines, removed 1 line │ ┃ Add a to-do… ⏎ Add ⏺ Bash(bun test) │ ┃ ⎿ 3 pass, 1 fail │ ┃ Inbox (0 queued) │ ┃ Queue tasks with /inbox <task>, then send ● Done. refresh now rejects expired claims and logs an audit event. │ ┃ them all at once. │ ┃ Queue a task for a subagent… ⏎ Add ✻ Worked for 42s · done 4:20 PM │ │ › /todo │ ⎿ task-inbox: No open to-dos. │ │ ────────────────────────────────────────────────────────────────────────────────────────────────────────────────────── › ? for shortcuts

Draws

Pane · Tasks
Claude's plan No plan yet. It fills in as Claude tracks its steps. My to-dos (0) Add a to-do… ⏎ Add Inbox (0 queued) Queue tasks with /inbox <task>, then send them all at once. Queue a task for a subagent… ⏎ Add
README

claude-mods

Claude Code mods (hooks-module plugins) for subagent visibility, task tracking, token savings and daily workflow. Needs Claude Code 2.1.304 or newer (built and tested on 2.1.304; the mod API is early access and can change between releases).

Install

At the prompt of a Claude Code terminal session:

/plugin install agent-monitor --marketplace samarthsh-git/claude-mods
/plugin install task-inbox --marketplace samarthsh-git/claude-mods
/plugin install token-saver --marketplace samarthsh-git/claude-mods
/plugin install daily-driver --marketplace samarthsh-git/claude-mods
/plugin install clarify-first --marketplace samarthsh-git/claude-mods
/plugin install mod-scout --marketplace samarthsh-git/claude-mods

Answer y to add the marketplace, then pick the user scope. From a shell: claude plugin install <mod> --marketplace samarthsh-git/claude-mods.

In the desktop app, where /plugin does not run: click + next to the prompt box, then Plugins, Add marketplace with samarthsh-git/claude-mods, then Add plugin.

If this repository is private, the machine needs a git credential for github.com (gh auth login, then gh auth setup-git); the install never prompts for one.

The mods

agent-monitor: what every subagent is doing

  • A sidebar (/subagents) that opens by itself when the first subagent starts.
  • Each subagent shows its task, type, model, run time, tool count and tokens.
  • A live "now:" line shows what it is doing at that moment ($ npm test, reading auth.ts, searching "TODO").
  • Finished agents show a one-line result. History stays across sessions (last 60).
  • Press a row to open that agent's transcript. /subagents clear wipes the history.
  • The status line shows ⚙ N subagents running, and a toast appears when one fails.

task-inbox: to-do list and task inbox

The Tasks pane (/todo or /inbox) has three parts:

  1. Claude's plan. A live mirror of Claude's own TodoWrite or Task list: ☐ pending, ◐ in progress, ☑ done.
  2. My to-dos. Your own list, saved per project. /todo <text> adds, /todo done 2 ticks, /todo rm 2 removes, /todo clear drops finished ones.
  3. Inbox. Queue tasks with /inbox <text>, then press Send (or /inbox send). Claude spawns one background subagent per task, in parallel, on the cheapest model that fits. You keep working while they run, and agent-monitor shows their progress.

token-saver: fewer tokens, same accuracy

FeatureWhat it doesSetting
Subagent model routingbalanced: search and lookup subagents (Explore, guides) run on Haiku. aggressive: general-purpose and Plan subagents spawned from an Opus-class session also run on Sonnet. A model Claude or you set explicitly is never changed, and neither are forks or plugin agents.routing
Lean-output rulesAdds about 150 tokens to the system prompt: no preamble or recaps, locate before reading, batch tool calls, delegate broad searches, prefer Edit, never trade correctness for brevity. It sits in the cached part of the prompt.leanRules
Duplicate-read guardRefuses once a Read of a file Claude already read with the same range in the same context while the file is unchanged on disk. Repeating the exact call goes through, so the worst case is one extra round trip. It resets on compaction.dedupeReads
Usage HUDStatus line: ctx 42% · $1.23 · 5h 37%. Toasts at 70% context (suggests /compact or /clear) and at 80% of a rate limit.always on
Model advisorA toast when a prompt looks trivial for an Opus-class model, or hard for Haiku. Pure regex, costs no tokens, at most once per 15 min.advisor
/tokensReport: fresh, cached and output tokens, cache hit rate, subagent share, what token-saver saved, and concrete tips.

Change settings with /config (or pluginConfigs in settings.json).

What saves the most, from experience: /clear between unrelated tasks, keeping the prompt cache warm (no model switches mid-thread), delegating read-heavy searches to subagents, and using Sonnet as the default with Opus only for hard problems. The mod nudges toward these; it cannot measure your exact savings.

daily-driver: get more done per day

  • /aim <goal> sets a focus goal (/aim auto <goal> sets one only when none is set), shown above the prompt with time spent and today's file and commit counts. Press Done or run /aim done to log it.
  • When a turn ran 60 s or longer (setting notifyAfterSeconds), you get a desktop or phone notification with Claude's answer. Hand Claude long work and switch tasks.
  • /eod (or /eod yesterday) prints a zero-token work log: Claude time, prompts, files changed, commits, test runs, subagents and goals done. Paste it into a standup.
  • /handoff has Claude write .claude/handoff.md (goal, done, state, next steps, gotchas). /pickup in the next session continues from it, and a toast reminds you when one exists.

clarify-first: questions before complex work

On a complex task or a deep, open-ended question, Claude first checks the repo and the chat for what it can learn itself. If something that would change the result is still missing, it asks you up to 4 multiple-choice questions in one round, with a recommended option on each. It then states the agreed spec in 3 lines or fewer, does the work, and checks the result against that spec. Everyday prompts ("fix the typo", "what is X") pass straight through.

Cost:

  • Detection: free keyword scoring decides most prompts. Only borderline ones get a classification call on the small, fast model, typically a few hundred tokens. Set detector to heuristic for zero detection tokens.
  • Instruction: about 200 tokens of instruction reach the model, and only on prompts flagged complex.
  • Payoff: one question round is usually cheaper than redoing work built on a wrong assumption. That trade is the point of the mod. It is not a measured gain.

Commands: /clarify on (default) · off · always · now <task> (force questions for one task) · /clarify alone shows status. Settings: detector (hybrid / heuristic / model), maxQuestions (1 to 4).

mod-scout: the right mod at the right time

It reads each prompt locally, with no model call and zero tokens, and never changes what Claude sees:

  • Suggests the installed mod that fits, as a toast, once per kind per session. For example "why is my usage so high" suggests /tokens, a list of 3 or more tasks suggests /inbox, and "wrapping up for today" suggests /handoff. It knows every mod here plus Podium and Grounded Engineering, and stays quiet about mods you have not installed.
  • Sets your focus goal (daily-driver's /aim) from your first real task of the session, unless you already set one.
  • Makes mods for you: send the same prompt a third time and it becomes a slash command on the spot (/run-test, say). This costs no tokens to make or to use beyond the prompt itself. /scout rm <name> removes one.
  • Builds bigger mods on request: /scout build <idea> has Claude write, validate and test a new mod, with zero run-time token cost wherever the feature allows. This is the one part that spends tokens, and only when you ask.

/scout shows recent suggestions and your macros. /scout off pauses it. Settings: autoMacros, autoFocus.

Claude chat (claude.ai, phone and desktop apps)

The mods above run only in Claude Code: the chat app has no plugin system. For chat, chat/ holds the closest equivalents, and both work on every device once set up:

  1. chat/preferences.txt: paste it into Settings → Profile → personal preferences. It holds the lean-output rules, clarifying questions on complex tasks only, and a model-fit hint. It is about 200 words, shorter than a typical preferences block, and preferences ride on every message, so shorter is cheaper.
  2. chat/deep-work-skill.zip: upload it in Settings → Capabilities → Skills. It is a gather → ask → spec → deliver → self-check procedure that Claude loads only for complex work, so everyday chats pay just its one-line description.

Chat habits that save the most: start a new chat per topic (each message resends the whole chat), put reference files in a Project instead of pasting them again, and use a lighter model for simple asks.

Cloud sessions from your phone

.claude/settings.json registers this marketplace and enables the six mods, so a Claude Code session opened on this repository (from claude.ai/code or the phone app) offers to install them with no commands typed.

Third-party mods from the same Reddit thread

These are other people's projects. Install them from their own repositories:

ModWhat it isInstall
PodiumHands jobs to other agent CLIs (Codex, Gemini, other Claude models) and tracks them in one pane, with sprints and PRs. Needs Node 22+, git 2.28+ and at least one other agent CLI./plugin marketplace add madjagstudios/plugins, then /plugin install podium@madjagstudios
Grounded Engineering/grounded pane: engineering practices that fit your repo, with sources, plus a reviewed shelf of skill repositories./plugin marketplace add madjagstudios/grounded-engineering, then /plugin install grounded-engineering@grounded-engineering
claude-ui-modNot a Claude Code plugin. It patches the files of the Claude Desktop app on Windows: subagent HUD, task rail, usage meters, provider routing.Windows only. Quit Claude Desktop, then git clone https://gitlab.com/blockshooter/claude-ui-mod && cd claude-ui-mod && npm ci && node scripts/verify-frontend.cjs && node patch.js apply. Read its SECURITY.md first: it rewrites your Claude Desktop install, and app updates can break it. node patch.js revert undoes it.

The "optimaizr" and "inbox monitor" mods mentioned in the thread had no links. task-inbox and token-saver cover the same ground.

Developing

claude plugin validate mods/<mod>
claude plugin test mods/<mod>

To run a working copy, add "env": { "CLAUDE_CODE_PLUGIN_DIRS": "/abs/path/mods/agent-monitor:/abs/path/mods/task-inbox" } to ~/.claude/settings.json, or start with claude --plugin-dir mods/<mod>.

License

MIT

Source 2 files
hooks/register.tsx 300 lines
1import { atom, read, update } from 'claude-code'
2import type { EngineInterface, Register } from 'claude-code'
3
4import type { InboxItem, PlanItem, Todo } from '../types'
5
6const PANE = 'task-inbox'
7const TITLE = 'Tasks'
8
9const plan = atom({ plugin: 'task-inbox', key: 'plan' } as const, [])
10const todos = atom({ plugin: 'task-inbox', key: 'todos' } as const, [])
11const inbox = atom({ plugin: 'task-inbox', key: 'inbox' } as const, [])
12
13function newId(now: number): string {
14  return `${now.toString(36)}${Math.floor(Math.random() * 1e6).toString(36)}`
15}
16
17function clip(text: string, n: number): string {
18  const one = text.replace(/\s+/g, ' ').trim()
19  return one.length > n ? one.slice(0, n - 1) + '…' : one
20}
21
22// The one prompt that hands the queued tasks to background subagents.
23function dispatchPrompt(items: readonly InboxItem[]): string {
24  const list = items.map((it, i) => `${i + 1}. ${it.text}`).join('\n')
25  return [
26    `Inbox dispatch (${items.length} task${items.length === 1 ? '' : 's'}). Run each task below in its own background subagent`,
27    '(Agent tool, run_in_background: true), all spawned in one message so they run in parallel.',
28    'Give each subagent a self-contained prompt with the context it needs.',
29    'Pick the cheapest model that will do the task well: haiku for search, lookup and summaries;',
30    'sonnet for routine, well-specified edits; leave the model unset for hard design or debugging.',
31    'Do not wait on them; when each finishes, report one line: task, outcome, and anything I must decide.',
32    '',
33    list,
34  ].join('\n')
35}
36
37async function persist($: EngineInterface, project: string) {
38  await $.store.set(`todos:${project}`, await read($, todos))
39  await $.store.set(`inbox:${project}`, await read($, inbox))
40}
41
42async function addTodo($: EngineInterface, project: string, text: string) {
43  const now = await $.clock.now()
44  await update($, todos, list => [...list, { id: newId(now), text, isDone: false, createdAt: now }])
45  await persist($, project)
46}
47
48async function addInbox($: EngineInterface, project: string, text: string) {
49  const now = await $.clock.now()
50  await update($, inbox, list => [...list, { id: newId(now), text, status: 'queued' as const, sentAt: null }])
51  await persist($, project)
52}
53
54async function dispatch($: EngineInterface, project: string, ids: readonly string[] | null): Promise<number> {
55  const list = await read($, inbox)
56  const picked = list.filter(it => it.status === 'queued' && (ids === null || ids.includes(it.id)))
57  if (picked.length === 0) return 0
58  // Submitted on a timer: a prompt cannot be submitted from inside the
59  // command or press that holds the session, only once it has returned.
60  const text = dispatchPrompt(picked)
61  $.clock.after(100, () => void $.prompt.submit({ text }))
62  const now = await $.clock.now()
63  const sent = new Set(picked.map(it => it.id))
64  await update($, inbox, l => l.map(it => (sent.has(it.id) ? { ...it, status: 'sent' as const, sentAt: now } : it)))
65  await persist($, project)
66  return picked.length
67}
68
69async function refreshStatus($: EngineInterface) {
70  const p = await read($, plan)
71  const mine = (await read($, todos)).filter(t => !t.isDone).length
72  const queued = (await read($, inbox)).filter(it => it.status === 'queued').length
73  const parts: string[] = []
74  if (p.length > 0) parts.push(`plan ${p.filter(x => x.status === 'completed').length}/${p.length}`)
75  if (mine > 0) parts.push(`${mine} to-do`)
76  if (queued > 0) parts.push(`${queued} in inbox`)
77  $.ui.status(parts.length > 0 ? `☑ ${parts.join(' · ')}` : undefined)
78}
79
80export const register: Register = on => {
81  let project = ''
82
83  on('session.start', async ($, e, next) => {
84    project = e.cwd
85    await $.command.register({
86      name: 'todo',
87      description: 'Your to-do list: /todo <text> adds, /todo done <n> ticks, /todo opens the Tasks pane',
88      argumentHint: '[text | done <n> | rm <n> | clear]',
89    })
90    await $.command.register({
91      name: 'inbox',
92      description: 'Queue tasks for background subagents: /inbox <text> queues, /inbox send dispatches all',
93      argumentHint: '[text | send | rm <n> | clear]',
94    })
95    const savedTodos = await $.store.get(`todos:${project}`)
96    const savedInbox = await $.store.get(`inbox:${project}`)
97    if (Array.isArray(savedTodos)) await update($, todos, () => savedTodos as Todo[])
98    if (Array.isArray(savedInbox)) await update($, inbox, () => savedInbox as InboxItem[])
99    await refreshStatus($)
100
101    return next(e)
102  })
103
104  on('command.run', { command: 'todo' }, async ($, e) => {
105    const args = e.args.trim()
106    const list = await read($, todos)
107    const open = list.filter(t => !t.isDone)
108    const m = /^(done|rm)\s+(\d+)$/.exec(args)
109    if (args === '') {
110      await $.ui.open({ id: PANE, title: TITLE })
111      return { text: open.length === 0 ? 'No open to-dos.' : open.map((t, i) => `${i + 1}. ${t.text}`).join('\n') }
112    }
113    if (args === 'clear') {
114      await update($, todos, l => l.filter(t => !t.isDone))
115      await persist($, project)
116      await refreshStatus($)
117      return { text: 'Cleared finished to-dos.' }
118    }
119    if (m) {
120      const target = open[Number(m[2]) - 1]
121      if (!target) return { text: `No open to-do #${m[2]}.` }
122      await update($, todos, l =>
123        m[1] === 'rm' ? l.filter(t => t.id !== target.id) : l.map(t => (t.id === target.id ? { ...t, isDone: true } : t)),
124      )
125      await persist($, project)
126      await refreshStatus($)
127      return { text: `${m[1] === 'rm' ? 'Removed' : 'Done'}: ${target.text}` }
128    }
129    await addTodo($, project, args)
130    await refreshStatus($)
131    return { text: `Added to-do #${open.length + 1}: ${args}` }
132  })
133
134  on('command.run', { command: 'inbox' }, async ($, e) => {
135    const args = e.args.trim()
136    const queued = (await read($, inbox)).filter(it => it.status === 'queued')
137    const m = /^rm\s+(\d+)$/.exec(args)
138    if (args === '') {
139      await $.ui.open({ id: PANE, title: TITLE })
140      return { text: queued.length === 0 ? 'Inbox is empty.' : queued.map((it, i) => `${i + 1}. ${it.text}`).join('\n') }
141    }
142    if (args === 'send') {
143      const n = await dispatch($, project, null)
144      await refreshStatus($)
145      return { text: n === 0 ? 'Nothing queued.' : `Dispatching ${n} task${n === 1 ? '' : 's'} to background subagents.` }
146    }
147    if (args === 'clear') {
148      await update($, inbox, l => l.filter(it => it.status === 'queued'))
149      await persist($, project)
150      return { text: 'Cleared sent tasks.' }
151    }
152    if (m) {
153      const target = queued[Number(m[1]) - 1]
154      if (!target) return { text: `No queued task #${m[1]}.` }
155      await update($, inbox, l => l.filter(it => it.id !== target.id))
156      await persist($, project)
157      await refreshStatus($)
158      return { text: `Removed: ${target.text}` }
159    }
160    await addInbox($, project, args)
161    await refreshStatus($)
162    return { text: `Queued #${queued.length + 1}: ${args}. /inbox send dispatches the queue.` }
163  })
164
165  // Mirror Claude's own plan (TodoWrite, or TaskCreate/TaskUpdate) from the main loop.
166  on('tool.call', { tool: 'TodoWrite' }, async ($, e, next) => {
167    const ran = await next(e)
168    if (e.agentId === undefined && ran.deny === undefined && ran.isError !== true) {
169      await update($, plan, () => e.todos.map((t, i): PlanItem => ({ id: String(i), text: t.content, status: t.status })))
170      await refreshStatus($)
171    }
172    return ran
173  }).catch(($, e, next) => next(e))
174
175  on('tool.call', { tool: 'TaskCreate' }, async ($, e, next) => {
176    const ran = await next(e)
177    if (e.agentId === undefined && ran.deny === undefined && ran.isError !== true) {
178      const id = ran.result.task.id
179      await update($, plan, l => [...l.filter(x => x.id !== id), { id, text: e.subject, status: 'pending' as const }])
180      await refreshStatus($)
181    }
182    return ran
183  }).catch(($, e, next) => next(e))
184
185  on('tool.call', { tool: 'TaskUpdate' }, async ($, e, next) => {
186    const ran = await next(e)
187    if (e.agentId === undefined && ran.deny === undefined && ran.isError !== true) {
188      const status = e.status
189      await update($, plan, l =>
190        status === 'deleted'
191          ? l.filter(x => x.id !== e.taskId)
192          : l.map(x => (x.id === e.taskId ? { ...x, text: e.subject ?? x.text, status: status ?? x.status } : x)),
193      )
194      await refreshStatus($)
195    }
196    return ran
197  }).catch(($, e, next) => next(e))
198
199  on('ui.render', { component: 'Pane', requestId: PANE }, async ($, e) => {
200    const els = $.ui.resolve(e)
201    const { Box, Text, Button } = els
202    const width = Math.max(20, e.props.bodyColumns ?? 40)
203    const p = await read($, plan)
204    const mine = await read($, todos)
205    const box = await read($, inbox)
206    const queued = box.filter(it => it.status === 'queued')
207    const sent = box.filter(it => it.status === 'sent')
208    const mark = { pending: '☐', in_progress: '◐', completed: '☑' } as const
209    const field = (key: string, placeholder: string, onSubmit: (text: string) => Promise<void>) =>
210      'Input' in els ? (
211        <els.Input key={key} placeholder={placeholder} submitLabel="Add" onSubmit={text => void (text.trim() && onSubmit(text.trim()))} />
212      ) : null
213
214    return (
215      <Box flexDirection="column">
216        <Text bold>Claude's plan {p.length > 0 ? `(${p.filter(x => x.status === 'completed').length}/${p.length})` : ''}</Text>
217        {p.length === 0 && <Text dimColor>No plan yet. It fills in as Claude tracks its steps.</Text>}
218        {p.map(x => (
219          <Text key={`p-${x.id}`} color={x.status === 'in_progress' ? 'claude' : undefined} dimColor={x.status === 'completed'} wrap="truncate">
220            {mark[x.status]} {clip(x.text, width - 3)}
221          </Text>
222        ))}
223
224        <Box marginTop={1}>
225          <Text bold>My to-dos ({mine.filter(t => !t.isDone).length})</Text>
226        </Box>
227        {mine.map(t => (
228          <Box key={`t-${t.id}`} flexDirection="row">
229            <Button
230              key={`tt-${t.id}`}
231              plain
232              onPress={async () => {
233                await update($, todos, l => l.map(x => (x.id === t.id ? { ...x, isDone: !x.isDone } : x)))
234                await persist($, project)
235                await refreshStatus($)
236              }}
237            >
238              {t.isDone ? '☑' : '☐'} <Text dimColor={t.isDone} strikethrough={t.isDone}>{clip(t.text, width - 8)}</Text>
239            </Button>
240            <Button
241              key={`tx-${t.id}`}
242              plain
243              dimColor
244              label=" ✕"
245              onPress={async () => {
246                await update($, todos, l => l.filter(x => x.id !== t.id))
247                await persist($, project)
248                await refreshStatus($)
249              }}
250            />
251          </Box>
252        ))}
253        {field('new-todo', 'Add a to-do…', async text => {
254          await addTodo($, project, text)
255          await refreshStatus($)
256        })}
257
258        <Box marginTop={1}>
259          <Text bold>Inbox ({queued.length} queued)</Text>
260        </Box>
261        {queued.length === 0 && <Text dimColor>Queue tasks with /inbox &lt;task&gt;, then send them all at once.</Text>}
262        {queued.map(it => (
263          <Box key={`i-${it.id}`} flexDirection="row">
264            <Text wrap="truncate">• {clip(it.text, width - 14)} </Text>
265            <Button key={`is-${it.id}`} label="Send" onPress={async () => (await dispatch($, project, [it.id]), refreshStatus($))} />
266            <Button
267              key={`ix-${it.id}`}
268              plain
269              dimColor
270              label=" ✕"
271              onPress={async () => {
272                await update($, inbox, l => l.filter(x => x.id !== it.id))
273                await persist($, project)
274                await refreshStatus($)
275              }}
276            />
277          </Box>
278        ))}
279        {field('new-inbox', 'Queue a task for a subagent…', async text => {
280          await addInbox($, project, text)
281          await refreshStatus($)
282        })}
283        {queued.length > 1 && (
284          <Button key="send-all" variant="primary" label={`Send all ${queued.length}`} onPress={async () => (await dispatch($, project, null), refreshStatus($))} />
285        )}
286        {sent.length > 0 && (
287          <Box flexDirection="column" marginTop={1}>
288            <Text dimColor>Sent ({sent.length}): progress shows in /subagents</Text>
289            {sent.slice(-5).map(it => (
290              <Text key={`s-${it.id}`} dimColor wrap="truncate">
291                ↗ {clip(it.text, width - 3)}
292              </Text>
293            ))}
294          </Box>
295        )}
296      </Box>
297    )
298  })
299}
300
types/index.d.ts 14 lines
1export type PlanStatus = 'pending' | 'in_progress' | 'completed'
2
3export type PlanItem = { id: string; text: string; status: PlanStatus }
4
5export type Todo = { id: string; text: string; isDone: boolean; createdAt: number }
6
7export type InboxItem = { id: string; text: string; status: 'queued' | 'sent'; sentAt: number | null }
8
9declare module 'claude-code' {
10  interface PluginState {
11    'task-inbox': { plan: PlanItem[]; todos: Todo[]; inbox: InboxItem[] }
12  }
13}
14