🌊 The original agent harness. Deploy intelligent multi-player swarms, coordinate autonomous workflows, and build conversational AI systems. Features adaptive memory, self-learning intelligence, federation, vector RAG integration, and native Claude Code / Codex / Hermes and many more Integrated
TypeScript agentic-aiagentic-frameworkagentic-workflowagentsai-agentsai-assistantai-skillsautonomous-agentsclaudecodexdsh-pluginharnessmcp-servermulti-agentmulti-agent-systemsnpmskillsswarmswarm-intelligencetypescript
feat(ruflo-console): Claude control confirms itself by default (autopilot inside the chosen level)
ruv5c1586d5946a
diff --git a/plugins/ruflo-console/hooks/help-topics-more.ts b/plugins/ruflo-console/hooks/help-topics-more.tsindex 6c4b88eba5..d132191c55 100644--- a/plugins/ruflo-console/hooks/help-topics-more.ts+++ b/plugins/ruflo-console/hooks/help-topics-more.ts@@ -54,13 +54,13 @@ export const MORE_TOPICS: readonly HelpTopic[] = [summary: 'Claude opens pages, fills fields and runs actions for you, as far as you allow',steps: [{ text: 'Choose how far in Settings → Claude control: read (look and open pages), write (also fill fields and run local actions), manage (also network), full (also spend, deploy, delete). It is off until you turn it on.', go: { view: 'settings' }, label: 'Open Settings' },- { text: 'Choose who says Yes in Settings → Claude control: confirm. ask (the default) leaves each action Claude requests waiting for your Yes in the console.' },- { text: 'auto lets Claude’s own call confirm an action, but only inside the level above: anything higher is still refused and nothing runs. Every auto action is logged on Overview.' },- { text: 'Start with ask. Use auto at write for local work you trust; use it at manage or full only when you are watching, because a deploy or a delete then runs without a Yes.' },+ { text: 'Choose who says Yes in Settings → Claude control: confirm. auto (the default) lets Claude run unattended inside the level; ask leaves each action waiting for your Yes in the console.' },+ { text: 'With auto, Claude’s own call confirms an action, but only inside the level above: anything higher is still refused and nothing runs. Every auto action is logged on Overview.' },+ { text: 'The level is your real limit: write for local work, manage adds the network, full adds spend, deploy and delete. Use manage or full only when you are watching, or set confirm to ask, because a deploy or a delete then runs without a Yes.' },{ text: 'Start a new Claude Code session (or /reload-plugins). Claude now has four tools: console_state, console_open, console_set and console_run. Ask in words: "set up a mission to add a dark mode toggle".' },{ text: 'Watch Overview → Claude control: the level, whether it asks or confirms itself, the count, and each action (done, waiting for you, refused). Take back control stops every tool at once; Give control back resumes.', go: { view: 'overview' }, label: 'Open Overview' },],- tips: ['auto-confirm is not Missions’ auto-run: auto-run hands tasks to Claude; auto-confirm lets Claude answer the console’s own Yes.', 'Text from peers, the web or files on screen is data to Claude, not instructions.'],+ tips: ['If Claude Code itself asks permission for a console tool, allow mcp__ruflo-console__* in your permissions settings: that prompt is Claude Code’s, not the console’s.', 'auto-confirm is not Missions’ auto-run: auto-run hands tasks to Claude; auto-confirm lets Claude answer the console’s own Yes.', 'Text from peers, the web or files on screen is data to Claude, not instructions.'],related: ['settings', 'mission', 'security'],keywords: ['claude', 'control', 'drive', 'computer', 'use', 'tools', 'autonomy', 'level', 'auto', 'ask', 'take', 'back', 'confirm', 'autoconfirm', 'approve', 'yes'],},diff --git a/plugins/ruflo-console/hooks/model-tools.ts b/plugins/ruflo-console/hooks/model-tools.tsindex f01bd4a77e..5d64ecb38c 100644--- a/plugins/ruflo-console/hooks/model-tools.ts+++ b/plugins/ruflo-console/hooks/model-tools.ts@@ -64,11 +64,11 @@ export function classOf(pending: Pick<Pending, 'label' | 'args' | 'note' | 'showexport const allows = (level: ControlLevel, kind: ActionClass): boolean => rank(level) >= rank(NEEDS[kind])export const levelOf = (value: unknown): ControlLevel => LEVELS.find(level => level === value) ?? 'off'-export const confirmOf = (value: unknown): ControlConfirm => (value === 'auto' ? 'auto' : 'ask')+export const confirmOf = (value: unknown): ControlConfirm => (value === 'ask' ? 'ask' : 'auto')/** `RUFLO_CONSOLE_CONTROL=write:auto` (level, then ask|auto): one session's setting, for a recording or a test; null when not a valid pair. */export function parseControlEnv(value: unknown): { level: ControlLevel; confirm: ControlConfirm } | null {- const [level, confirm = 'ask'] = typeof value === 'string' ? value.trim().toLowerCase().split(':') : []+ const [level, confirm = 'auto'] = typeof value === 'string' ? value.trim().toLowerCase().split(':') : []const found = LEVELS.find(candidate => candidate === level)return found === undefined || (confirm !== 'ask' && confirm !== 'auto') ? null : { level: found, confirm }diff --git a/plugins/ruflo-console/hooks/settings.ts b/plugins/ruflo-console/hooks/settings.tsindex 5006d8ad9b..5fe75a591e 100644--- a/plugins/ruflo-console/hooks/settings.ts+++ b/plugins/ruflo-console/hooks/settings.ts@@ -46,7 +46,7 @@ export const CLAUDE_MODELS = ['default', 'haiku', 'sonnet', 'opus'] as constexport const AI_BUDGETS = [0.25, 0.5, 1, 2] as const/** `autoAccept`: claude, codex and swarm turns go straight out with no confirm (they stay read-only, in plan mode, under the budget); ruflo commands still ask. */export type AiPrefs = { claudeModel: (typeof CLAUDE_MODELS)[number]; budgetUsd: (typeof AI_BUDGETS)[number]; autoAccept: boolean; /** Claude writes guidance after a mission goal is entered. */ guidance: boolean; /** ADR-443: the active mission and task ride in Claude's prompt (changes only when the task changes). */ missionContext: boolean; /** ADR-443: the person's own gate commands, one per line or `\n`-separated (parseGates validates). */ loopGates: string; /** ADR-443: a USD cap for one mission's spend ('' none). */ missionCapUsd: string; /** ADR-444: how far Claude may drive the console with its console_* tools (they exist only when not off). */ modelControl: 'off' | 'read' | 'write' | 'manage' | 'full'; /** ADR-444: a non-read action waits for the person's Yes (ask) or confirms itself (auto). */ modelConfirm: 'ask' | 'auto' } & LoopPrefs-export const DEFAULT_AI: AiPrefs = { claudeModel: 'default', budgetUsd: 1, autoAccept: false, guidance: true, missionContext: true, loopGates: '', missionCapUsd: '', modelControl: 'off', modelConfirm: 'ask', ...DEFAULT_LOOP }+export const DEFAULT_AI: AiPrefs = { claudeModel: 'default', budgetUsd: 1, autoAccept: false, guidance: true, missionContext: true, loopGates: '', missionCapUsd: '', modelControl: 'off', modelConfirm: 'auto', ...DEFAULT_LOOP }const ON_OFF = ['on', 'off'] as constconst onOff = (value: boolean) => (value ? 'on' : 'off')@@ -283,7 +283,7 @@ export async function loadAiPrefs(state: State, host: Host): Promise<void> {autoAccept: stored?.autoAccept === true,guidance: stored?.guidance !== false,modelControl: (['read', 'write', 'manage', 'full'] as const).find(item => item === stored?.modelControl) ?? 'off',- modelConfirm: stored?.modelConfirm === 'auto' ? 'auto' : 'ask',+ modelConfirm: stored?.modelConfirm === 'ask' ? 'ask' : 'auto',missionContext: stored?.missionContext !== false,loopGates: typeof stored?.loopGates === 'string' ? stored.loopGates.slice(0, 800) : '',missionCapUsd: typeof stored?.missionCapUsd === 'string' && /^\d{1,5}(\.\d{1,2})?$/.test(stored.missionCapUsd) ? stored.missionCapUsd : '',diff --git a/plugins/ruflo-console/hooks/views/settings.ts b/plugins/ruflo-console/hooks/views/settings.tsindex ee771cff19..78770fae7b 100644--- a/plugins/ruflo-console/hooks/views/settings.ts+++ b/plugins/ruflo-console/hooks/views/settings.ts@@ -161,7 +161,7 @@ function aiItems(ctx: Ctx): Item[] {one('guidance', 'Mission guidance', 'after a mission goal is entered, claude -p writes detailed guidance by lifecycle stage and suggests ruflo capabilities to bring in (it asks first unless always accept)', ai.guidance ? 'on' : 'off', ['on', 'off'], !ai.guidance, value => ctx.act.settings.ai({ guidance: value === 'on' }), 'mission goal guidance advice suggestions'),...LOOP_ROWS.map(row => one(row.id, row.title, row.description, row.current(ai), row.options, row.isChanged(ai), value => ctx.act.settings.ai(row.patch(value)), row.extra)),one('model-control', 'Claude control', 'how far Claude may drive this console with its console_* tools: off, read (look and open pages), write (also fill fields and run local actions), manage (also network), full (also spend, deploy, delete); takes effect in a new session or /reload-plugins', ai.modelControl, ['off', 'read', 'write', 'manage', 'full'], ai.modelControl !== 'off', value => ctx.act.settings.ai({ modelControl: value as typeof ai.modelControl }), 'claude control drive computer use tools autonomy model'),- one('model-confirm', 'Claude control: confirm', 'ask leaves each action Claude requests waiting for your Yes in the console; auto lets Claude’s call confirm itself, within the level above (every call is logged on Overview)', ai.modelConfirm, ['ask', 'auto'], ai.modelConfirm !== 'ask', value => ctx.act.settings.ai({ modelConfirm: value === 'auto' ? 'auto' : 'ask' }), 'claude control confirm auto ask approve'),+ one('model-confirm', 'Claude control: confirm', 'auto (the default) lets Claude’s call confirm itself, within the level above, so it runs unattended (every call is logged on Overview); ask leaves each action waiting for your Yes in the console', ai.modelConfirm, ['ask', 'auto'], ai.modelConfirm !== 'auto', value => ctx.act.settings.ai({ modelConfirm: value === 'auto' ? 'auto' : 'ask' }), 'claude control confirm auto ask approve'),one('ctx-mission', 'Mission context in Claude’s prompt', 'the active mission and task ride in Claude’s system prompt, and change only when the task does (a changed prompt makes Claude re-read the chat)', ai.missionContext ? 'on' : 'off', ['on', 'off'], !ai.missionContext, value => ctx.act.settings.ai({ missionContext: value === 'on' }), 'mission context prompt cache claude'),one('loop-gates', 'Mission gates', 'your own commands a mission may run to verify a task, one per line; each asks first and shows its exact argv; no shell characters', ai.loopGates, [], ai.loopGates !== '', value => ctx.act.settings.ai({ loopGates: value.slice(0, 800) }), 'gates verify tests smoke evidence'),one('mission-cap', 'Mission spend cap (USD)', 'auto-run pauses when one mission’s spend reaches this (list-price estimate; empty means no cap)', ai.missionCapUsd, [], ai.missionCapUsd !== '', value => ctx.act.settings.ai({ missionCapUsd: /^\d{1,5}(\.\d{1,2})?$/.test(value.trim()) ? value.trim() : '' }), 'mission cap budget spend cost'),diff --git a/plugins/ruflo-console/tests/model-tools.spec.ts b/plugins/ruflo-console/tests/model-tools.spec.tsindex 5ee3cbebf5..e278eecee6 100644--- a/plugins/ruflo-console/tests/model-tools.spec.ts+++ b/plugins/ruflo-console/tests/model-tools.spec.ts@@ -65,7 +65,9 @@ describe('levels, classes and the environment override', () => {expect(levelOf('root')).toBe('off')expect(levelOf(undefined)).toBe('off')expect(confirmOf('auto')).toBe('auto')- expect(confirmOf('whatever')).toBe('ask')+ expect(confirmOf('ask')).toBe('ask')+ expect(confirmOf(undefined)).toBe('auto')+ expect(confirmOf('whatever')).toBe('auto')expect([allows('read', 'read'), allows('read', 'write'), allows('write', 'write'), allows('write', 'network'), allows('manage', 'network'), allows('manage', 'spend'), allows('full', 'delete'), allows('off', 'read')]).toEqual([true, false, true, false, true, false, true, false])})@@ -88,7 +90,8 @@ describe('levels, classes and the environment override', () => {it('takes RUFLO_CONSOLE_CONTROL as level and confirm, or nothing', () => {expect(parseControlEnv('write:auto')).toEqual({ level: 'write', confirm: 'auto' })- expect(parseControlEnv('read')).toEqual({ level: 'read', confirm: 'ask' })+ expect(parseControlEnv('read')).toEqual({ level: 'read', confirm: 'auto' })+ expect(parseControlEnv('write:ask')).toEqual({ level: 'write', confirm: 'ask' })for (const bad of [undefined, '', 'god', 'write:maybe', 7]) expect(parseControlEnv(bad)).toBeNull()})})diff --git a/v3/docs/adr/ADR-444-claude-controls-the-console.md b/v3/docs/adr/ADR-444-claude-controls-the-console.mdindex c848ab274a..719d5a69f8 100644--- a/v3/docs/adr/ADR-444-claude-controls-the-console.md+++ b/v3/docs/adr/ADR-444-claude-controls-the-console.md@@ -35,7 +35,7 @@ Four tools, `mcp__ruflo-console__` plus:| `manage` | also run entries that touch the network (install, update, push) || `full` | also run entries that spend money or delete or stop things (deploys, cancellations, resets) |-and a second setting, **confirm**: `ask` (default) leaves every non-read action pending in the console's own confirm row for the person to answer; `auto` lets Claude's call confirm itself, within the level. An entry's class is read from its spec: the console's own read-only flag, else its label, command and notes. Words for deleting, stopping, resetting, cleaning up or migrating make it `delete`; words for a billed turn or "costs money" make it `spend`; words for the network (install, push, fetch, update, publish) make it `network`; anything else that is not read-only is a local `write`. The notes also say what an action does *not* do ("spends nothing", "not a charge", "runs no agent"), and those do not count.+and a second setting, **confirm**: `auto` (default since console 0.32: once the person has chosen a level, Claude runs unattended inside it) lets Claude's call confirm itself, within the level; `ask` leaves every non-read action pending in the console's own confirm row for the person to answer. An entry's class is read from its spec: the console's own read-only flag, else its label, command and notes. Words for deleting, stopping, resetting, cleaning up or migrating make it `delete`; words for a billed turn or "costs money" make it `spend`; words for the network (install, push, fetch, update, publish) make it `network`; anything else that is not read-only is a local `write`. The notes also say what an action does *not* do ("spends nothing", "not a charge", "runs no agent"), and those do not count.**The cockpit is the dashboard.** An "Claude control" section on Overview shows the level and confirm mode, whether control is on or paused, the call count, and a log of the last actions with their outcome (ok, waiting for you, refused, failed). A **Take back control** button pauses every tool at once (the next call is refused with that reason) and **Give control back** resumes. The log is the audit trail for a session.@@ -53,3 +53,7 @@ and a second setting, **confirm**: `ask` (default) leaves every non-read action- A separate human-confirm channel through Claude Code's permission dialog for `ask` mode.- Per-entry allow lists; the level is coarse by design.- Driving panes other than the console's own.++## 6. Amendment: autopilot by default++With `ask` as the default, a person who had turned control on found Claude parked on every write, waiting for a Yes that nothing prompted for. The level is the real limit (it defaults to off, and only the person raises it), so the confirm default is now `auto`; `ask` remains an explicit choice. Claude Code's own permission prompt for the four tools is separate: allow `mcp__ruflo-console__*` in the permissions settings to remove that one too (a mod can only tighten `tool.check`, never loosen it).