claude-code-kanban 5.3.0 → 5.4.0

package/plugin/plugins/claude-code-kanban/tests/activity.test.ts

@@ -0,0 +1,300 @@
+import type { On, ToolCallArgs, ToolCheckDecision } from 'claude-code'
+import { type Engine, expect, type MockClock, mock, test } from 'claude-code/testing'
+import { BOARD_ALLOWED, CLEARED } from '../hooks/activity'
+
+const ROOT = 'C:/cfg/.cck/agent-activity'
+const SESSION = `${ROOT}/sid-1`
+const MARKER = `${SESSION}/_waiting.json`
+const CONFIG = 'C:/cfg/.cck/config.json'
+
+type EngineOptions = { decision?: ToolCheckDecision; rule?: string; env?: Record<string, string> }
+
+// A test hook cannot raise tool.check, so the test raises it (askFor), and the engine's tool.call
+// holds the first call until the test answers the terminal prompt.
+function engine(on: On, files: Record<string, string>, { decision = 'allow', rule, env = {} }: EngineOptions = {}) {
+  const writes: { path: string; text: string }[] = []
+  const runs: unknown[] = []
+  const status: (string | undefined)[] = []
+  let answer = () => {}
+  const terminal = new Promise<void>(resolve => {
+    answer = resolve
+  })
+  const norm = (p: string) => p.replaceAll('\\', '/')
+  on('fs.read', (_$, e) => {
+    const text = files[norm(e.path)]
+    if (text === undefined) throw new Error(`ENOENT: ${e.path}`)
+    return { value: text }
+  })
+  on('fs.write', (_$, e) => {
+    writes.push({ path: norm(e.path), text: e.text })
+    files[norm(e.path)] = e.text
+    return { value: undefined }
+  })
+  on('env.get', (_$, e) => ({ value: { CLAUDE_CONFIG_DIR: 'C:/cfg', ...env }[e.name] }))
+  on('session.id', () => ({ value: 'sid-1' }))
+  on('session.cwd', () => ({ value: 'C:/proj' }))
+  on('ui.status', (_$, e) => {
+    status.push(e.text)
+    return { value: undefined }
+  })
+  on('tool.check', () => ({ decision, rule }))
+  let held = decision === 'ask'
+  on('tool.call', async (_$, e) => {
+    const { tool, tool_use_id, agentId, ...input } = e
+    if (held) {
+      held = false
+      await terminal
+    }
+    runs.push(input)
+    return { result: { stdout: 'ok' } }
+  })
+  on('session.start', (_$, e) => ({ cwd: e.cwd }))
+  on('agent.spawn', () => ({ model: 'haiku', agentId: 'a1' }))
+  on('turn.complete', (_$, e) => ({ text: e.answer }))
+  on('classic.*', () => ({}))
+  return { writes, runs, status, answer }
+}
+
+async function askFor($: Engine, clock: MockClock, args: { tool: string } & Record<string, unknown>) {
+  const { tool, ...input } = args
+  const call = $.tool.call(args as ToolCallArgs)
+  await clock.settle()
+  await $.tool.check({ tool, input })
+  await clock.settle()
+  return { call }
+}
+
+const marker = (files: Record<string, string>) => JSON.parse(files[MARKER]!)
+const decide = (files: Record<string, string>, decision: object) => {
+  files[`${SESSION}/_decision-${marker(files).id}.json`] = JSON.stringify(decision)
+}
+
+const lines = (text = '') => text.trim().split('\n').map(l => JSON.parse(l))
+
+const turn = (fields: { agentId?: string; answer?: string; isAborted?: boolean }) => ({
+  answer: '',
+  durationMs: 1,
+  isAborted: false,
+  turnId: 't',
+  reason: 'answer' as const,
+  ...fields,
+})
+
+const spawn = (subagentType: string, name?: string) => ({
+  tool_use_id: 'tu',
+  prompt: 'p',
+  description: 'd',
+  subagentType,
+  provider: { plugin: 'engine', tier: 'core' as const },
+  parentModel: 'opus',
+  background: false,
+  fork: false,
+  name,
+})
+
+test('an allowed tool call writes nothing', async ($, on) => {
+  const { writes } = engine(on, {})
+
+  const result = await $.tool.call({ tool: 'Bash', command: 'ls' })
+
+  expect(result.result).toEqual({ stdout: 'ok' })
+  expect(writes).toEqual([])
+})
+
+test('an ask writes the marker, and a terminal answer clears it', async ($, on) => {
+  const files: Record<string, string> = {}
+  const clock = mock.clock(on)
+  const { runs, answer } = engine(on, files, { decision: 'ask', rule: 'Bash(ls)' })
+
+  const { call } = await askFor($, clock, { tool: 'Bash', command: 'ls' })
+  expect(marker(files)).toMatchObject({
+    status: 'waiting',
+    kind: 'permission',
+    toolName: 'Bash',
+    toolInput: '{"command":"ls"}',
+    cwd: 'C:/proj',
+    rule: 'Bash(ls)',
+  })
+  await clock.advance(1500)
+  answer()
+
+  expect((await call).result).toEqual({ stdout: 'ok' })
+  expect(files[MARKER]).toBe(CLEARED)
+  expect(runs).toEqual([{ command: 'ls' }])
+})
+
+test('a board allow runs the tool with the board input and shows why', async ($, on) => {
+  const files: Record<string, string> = {}
+  const clock = mock.clock(on)
+  const { runs, status } = engine(on, files, { decision: 'ask' })
+
+  const { call } = await askFor($, clock, { tool: 'Bash', command: 'ls' })
+  decide(files, { behavior: 'allow', updatedInput: { command: 'ls -a' } })
+  await clock.advance(500)
+
+  expect((await call).result).toEqual({ stdout: 'ok' })
+  expect(runs).toEqual([{ command: 'ls -a' }])
+  expect(status).toEqual([BOARD_ALLOWED, undefined])
+  expect(files[MARKER]).toBe(CLEARED)
+})
+
+test('a board deny refuses the call with its message', async ($, on) => {
+  const files: Record<string, string> = {}
+  const clock = mock.clock(on)
+  const { runs } = engine(on, files, { decision: 'ask' })
+
+  const { call } = await askFor($, clock, { tool: 'Bash', command: 'rm -rf x' })
+  decide(files, { behavior: 'deny', message: 'not that' })
+  await clock.advance(500)
+
+  expect((await call).deny).toBe('not that')
+  expect(runs).toEqual([])
+  expect(files[MARKER]).toBe(CLEARED)
+})
+
+test('board answers to a question come back as the tool result', async ($, on) => {
+  const files: Record<string, string> = {}
+  const clock = mock.clock(on)
+  engine(on, files, { decision: 'ask' })
+  const questions = [{ question: 'Fruit?', header: 'Fruit', options: [], multiSelect: false }]
+
+  const { call } = await askFor($, clock, { tool: 'AskUserQuestion', questions })
+  expect(marker(files).kind).toBe('question')
+  decide(files, { answers: { 'Fruit?': 'Pear' } })
+  await clock.advance(500)
+
+  expect((await call).result).toEqual({ questions, answers: { 'Fruit?': 'Pear' } })
+})
+
+test('with approvals off the board only shows the ask', async ($, on) => {
+  const files: Record<string, string> = { [CONFIG]: '{"approvals":{"enabled":false}}' }
+  const clock = mock.clock(on)
+  const { runs, answer } = engine(on, files, { decision: 'ask' })
+
+  const { call } = await askFor($, clock, { tool: 'Bash', command: 'ls' })
+  decide(files, { behavior: 'deny' })
+  await clock.advance(2000)
+  answer()
+
+  expect((await call).result).toEqual({ stdout: 'ok' })
+  expect(runs).toEqual([{ command: 'ls' }])
+})
+
+test('the board stops waiting after waitSeconds', async ($, on) => {
+  const files: Record<string, string> = { [CONFIG]: '{"approvals":{"waitSeconds":1}}' }
+  const clock = mock.clock(on)
+  const { runs, answer } = engine(on, files, { decision: 'ask' })
+
+  const { call } = await askFor($, clock, { tool: 'Bash', command: 'ls' })
+  await clock.advance(1500)
+  decide(files, { behavior: 'deny' })
+  await clock.advance(1000)
+  answer()
+
+  expect((await call).result).toEqual({ stdout: 'ok' })
+  expect(runs).toEqual([{ command: 'ls' }])
+})
+
+test('a marker that a newer ask owns stays', async ($, on) => {
+  const files: Record<string, string> = {}
+  const clock = mock.clock(on)
+  const { answer } = engine(on, files, { decision: 'ask' })
+
+  const { call } = await askFor($, clock, { tool: 'Bash', command: 'ls' })
+  files[MARKER] = '{"status":"waiting","id":"newer"}'
+  answer()
+  await call
+
+  expect(files[MARKER]).toBe('{"status":"waiting","id":"newer"}')
+})
+
+test('a finished main turn marks the session unread', async ($, on) => {
+  const files: Record<string, string> = {}
+  engine(on, files)
+
+  await $.turn.complete(turn({}))
+
+  expect(files[`${SESSION}/_stop.json`]).toBe('')
+})
+
+test('an interrupted main turn does not mark the session unread', async ($, on) => {
+  const { writes } = engine(on, {})
+
+  await $.turn.complete(turn({ isAborted: true }))
+
+  expect(writes).toEqual([])
+})
+
+test('session start maps the session to a custom task list', async ($, on) => {
+  const map = `${ROOT}/_task-maps/list-1.json`
+  const files: Record<string, string> = { [map]: '{"old":{"project":"C:/x"}}' }
+  engine(on, files, { env: { CLAUDE_CODE_TASK_LIST_ID: 'list-1' } })
+
+  await $.session.start({ cwd: 'C:/proj', surface: null, isInteractive: false })
+
+  const maps = JSON.parse(files[map]!)
+  expect(maps.old).toEqual({ project: 'C:/x' })
+  expect(maps['sid-1'].project).toBe('C:/proj')
+})
+
+test('session start writes nothing without a custom task list', async ($, on) => {
+  const { writes } = engine(on, {})
+
+  await $.session.start({ cwd: 'C:/proj', surface: null, isInteractive: false })
+
+  expect(writes).toEqual([])
+})
+
+test('a subagent spawn then finish records start and stop', async ($, on) => {
+  const files: Record<string, string> = {}
+  engine(on, files)
+
+  await $.agent.spawn(spawn('general-purpose'))
+  await $.turn.complete(turn({ agentId: 'a1', answer: 'Task done with "quotes"' }))
+
+  const [start, stop] = lines(files[`${SESSION}/a1.jsonl`])
+  expect(start).toMatchObject({ agentId: 'a1', type: 'general-purpose', event: 'start', status: 'active' })
+  expect(stop).toMatchObject({ agentId: 'a1', event: 'stop', status: 'stopped', lastMessage: 'Task done with "quotes"' })
+  expect('type' in stop).toBe(false)
+  expect(files[`${SESSION}/_name-general-purpose.id`]).toBe('a1')
+  expect(files[`${SESSION}/_stop.json`]).toBeUndefined()
+})
+
+test('a subagent stop with an empty answer leaves lastMessage out', async ($, on) => {
+  const files: Record<string, string> = {}
+  engine(on, files)
+
+  await $.agent.spawn(spawn('general-purpose'))
+  await $.turn.complete(turn({ agentId: 'a1' }))
+
+  const [, stop] = lines(files[`${SESSION}/a1.jsonl`])
+  expect(stop).toMatchObject({ agentId: 'a1', event: 'stop' })
+  expect('lastMessage' in stop).toBe(false)
+})
+
+test('a finished loop that was never spawned writes nothing', async ($, on) => {
+  const { writes } = engine(on, {})
+
+  await $.turn.complete(turn({ agentId: 'internal' }))
+
+  expect(writes).toEqual([])
+})
+
+test('a named spawn maps the name, and TeammateIdle finds the agent by it', async ($, on) => {
+  const files: Record<string, string> = {}
+  engine(on, files)
+
+  await $.agent.spawn(spawn('general-purpose', 'researcher'))
+  await $.classic.TeammateIdle({ teammate_name: 'researcher', team_name: 'team' })
+
+  const idle = lines(files[`${SESSION}/a1.jsonl`])[1]
+  expect(idle).toMatchObject({ agentId: 'a1', type: 'researcher', event: 'idle', status: 'idle' })
+})
+
+test('TeammateIdle for an unknown teammate writes nothing', async ($, on) => {
+  const { writes } = engine(on, {})
+
+  await $.classic.TeammateIdle({ teammate_name: 'ghost', team_name: 'team' })
+
+  expect(writes).toEqual([])
+})

package/plugin/plugins/claude-code-kanban/tests/context.test.ts

@@ -0,0 +1,72 @@
+import type { On, SessionMeasureInput } from 'claude-code'
+import { expect, test } from 'claude-code/testing'
+import { displayName } from '../hooks/context'
+
+const MEASURE: SessionMeasureInput = {
+  context: { tokens: 129_174, window: 1_000_000, percent: 13 },
+  rateLimits: [
+    { kind: 'five_hour', percentUsed: 1, resetsAt: '2026-10-02T13:10:00Z' },
+    { kind: 'seven_day', percentUsed: 7.5, resetsAt: '2026-10-09T00:00:00Z' },
+  ],
+  cost: { usd: 2.71 },
+  changed: ['context', 'rateLimits', 'cost'],
+}
+
+function engine(on: On, env: Record<string, string>, sessionId: string) {
+  const writes: { path: string; text: string }[] = []
+  on('fs.write', (_$, e) => {
+    writes.push({ path: e.path.replaceAll('\\', '/'), text: e.text })
+    return { value: undefined }
+  })
+  on('env.get', (_$, e) => ({ value: env[e.name] }))
+  on('session.id', () => ({ value: sessionId }))
+  on('session.model', () => ({ value: 'claude-opus-5-5' }))
+  on('session.measure', (_$, e) => ({ changed: [...e.changed] }))
+  return writes
+}
+
+test('session.measure writes the statusLine shape the board reads', async ($, on) => {
+  const writes = engine(on, { CLAUDE_CONFIG_DIR: 'C:/cfg' }, 'sid-1')
+
+  await $.session.measure(MEASURE)
+
+  expect(writes.length).toBe(1)
+  expect(writes[0]?.path).toBe('C:/cfg/.cck/context-status/sid-1.json')
+  const status = JSON.parse(writes[0]?.text ?? '{}')
+  expect(status.model).toEqual({ id: 'claude-opus-5-5', display_name: 'Opus 5.5' })
+  expect(status.cost.total_cost_usd).toBe(2.71)
+  expect(status.context_window).toEqual({
+    total_input_tokens: 129_174,
+    context_window_size: 1_000_000,
+    current_usage: null,
+    used_percentage: 13,
+  })
+  expect(status.rate_limits.five_hour).toEqual({ used_percentage: 1, resets_at: 1790946600 })
+  expect(status.rate_limits.seven_day.used_percentage).toBe(7.5)
+})
+
+test('skips a write when the status has not changed', async ($, on) => {
+  const writes = engine(on, { CLAUDE_CONFIG_DIR: 'C:/cfg' }, 'sid-3')
+
+  await $.session.measure(MEASURE)
+  await $.session.measure(MEASURE)
+  await $.session.measure({ ...MEASURE, cost: { usd: 3 } })
+
+  expect(writes.length).toBe(2)
+})
+
+test('display name from model id', async () => {
+  expect(displayName('claude-opus-5-5')).toBe('Opus 5.5')
+  expect(displayName('claude-haiku-4-5-20251001')).toBe('Haiku 4.5')
+  expect(displayName('claude-fable-5-1[1m]')).toBe('Fable 5.1')
+  expect(displayName('claude-opus-4-20250514')).toBe('Opus 4')
+  expect(displayName('gpt-x')).toBe('gpt-x')
+})
+
+test('falls back to ~/.claude when CLAUDE_CONFIG_DIR is unset', async ($, on) => {
+  const writes = engine(on, { USERPROFILE: 'C:/Users/me' }, 'sid-2')
+
+  await $.session.measure(MEASURE)
+
+  expect(writes.map(w => w.path)).toEqual(['C:/Users/me/.claude/.cck/context-status/sid-2.json'])
+})

package/plugin/plugins/claude-code-kanban/tsconfig.json

@@ -0,0 +1,3 @@
+{
+  "extends": "./.claude-plugin/types/tsconfig.json"
+}

package/public/app.js

@@ -1541,7 +1541,7 @@ function renderWaitingBody(tool, params) {
 const PLAN_REJECT_MSG = 'Plan rejected from the kanban board.';
 
 // Answerable only when the marker carries a request id (written by
-// approval-gate.sh, which is polling for a decision) and the gate hasn't
+// the plugin's mod, which is polling for a decision) and the gate hasn't
 // lapsed past its wait window — older or lapsed markers are only answerable
 // in the terminal.
 function isWaitingAnswerable() {
@@ -1600,9 +1600,9 @@ function rejectWaitingPlan(inputId) {
   respondWaiting({ behavior: 'deny', message: feedback || PLAN_REJECT_MSG });
 }
 
-// UI-driven approvals: POST the decision the blocking approval-gate.sh hook is
-// polling for. First writer wins (D5) — a 409/410 means the ask was answered in
-// the terminal or superseded, so just drop the card.
+// UI-driven approvals: POST the decision the plugin's mod is polling for.
+// First writer wins (D5) — a 409/410 means the ask was answered in the
+// terminal or superseded, so just drop the card.
 async function respondWaiting(payload) {
   if (!currentSessionId || !currentWaiting?.id) return;
   try {
@@ -8764,7 +8764,7 @@ function renderContextDetail(session, { tokens = true } = {}) {
             <div class="stat-item"><span class="stat-label">Current output</span><span class="stat-value">${formatTokens((usage.output_tokens || 0) / 1000)}</span></div>
             <div class="stat-divider"></div>
             <div class="stat-item"><span class="stat-label">Total input</span><span class="stat-value">${formatTokens(ctx.inputTokens / 1000)}</span></div>
-            <div class="stat-item"><span class="stat-label">Total output</span><span class="stat-value">${formatTokens(ctx.outputTokens / 1000)}</span></div>
+            ${ctx.outputTokens ? `<div class="stat-item"><span class="stat-label">Total output</span><span class="stat-value">${formatTokens(ctx.outputTokens / 1000)}</span></div>` : ''}
             <div class="stat-divider"></div>`
     : '';
 
@@ -8784,9 +8784,9 @@ function renderContextDetail(session, { tokens = true } = {}) {
           <div class="detail-context-stats">
             ${tokenRows}
             <div class="stat-item"><span class="stat-label">Cost</span><span class="stat-value" style="color:${getCostColor(cost.total_cost_usd)}">${formatCost(cost.total_cost_usd)}</span></div>
-            <div class="stat-item"><span class="stat-label">Duration</span><span class="stat-value">${formatDuration(cost.total_duration_ms)}</span></div>
-            <div class="stat-item"><span class="stat-label">API time</span><span class="stat-value">${formatDuration(cost.total_api_duration_ms)}</span></div>
-            <div class="stat-item"><span class="stat-label">Lines</span><span class="stat-value"><span style="color:${CONTEXT_COLORS.green}">+${(cost.total_lines_added || 0).toLocaleString()}</span> / <span style="color:${CONTEXT_COLORS.red}">-${(cost.total_lines_removed || 0).toLocaleString()}</span></span></div>
+            ${cost.total_duration_ms != null ? `<div class="stat-item"><span class="stat-label">Duration</span><span class="stat-value">${formatDuration(cost.total_duration_ms)}</span></div>` : ''}
+            ${cost.total_api_duration_ms != null ? `<div class="stat-item"><span class="stat-label">API time</span><span class="stat-value">${formatDuration(cost.total_api_duration_ms)}</span></div>` : ''}
+            ${cost.total_lines_added != null ? `<div class="stat-item"><span class="stat-label">Lines</span><span class="stat-value"><span style="color:${CONTEXT_COLORS.green}">+${(cost.total_lines_added || 0).toLocaleString()}</span> / <span style="color:${CONTEXT_COLORS.red}">-${(cost.total_lines_removed || 0).toLocaleString()}</span></span></div>` : ''}
           </div>
         </div>`;
 }