Settings explorer

Hide Projects Sidebar

Collapse the projects sidebar to maximize chat area. A thin edge gutter remains clickable to re-open it. Shortcut: Ctrl+\

Hide Sessions Sidebar

Collapse the sessions sidebar to maximize chat area. A thin edge gutter remains clickable to re-open it. Shortcut: Ctrl+Shift+\

Auto-Detect Project Icons

Scan project folders for favicons and logos to display in the sidebar

Show current git branch in header

Display the active git branch next to the session name. Can be disabled per project.

Show structured agent prose inline

Show the agent’s in-between commentary (headings, lists, longer prose) as regular message bubbles instead of folding it into the activity row.

Theme

Dark, light, or system theme

Language

Choose the display language for the interface

Visual Theme

Choose a dark-mode skin for panels and accents

Sidebar Position

Place project and session sidebars on the left or right side of the window

Font Size

Adjust text size on desktop (phones use the separate Mobile text size)

Mobile text size

Text size on phones and narrow touch screens, kept separate from the desktop size

Custom Themes

Import or create custom visual themes from .theme.json files

Chat Depth

How much shadow and glassy depth the chat messages have. Flat is the classic look; Glass is the most three-dimensional.

Motion & Polish

Optional visual flourishes, depth, hover lift, count-up stats, ambient backgrounds. Off by default; a master switch plus per-effect toggles.

Mind Map

Default color mode and visual theme for every mind map (each map can override it)

Track my typing speed

Measure your real typing speed in the message composer and show it in Statistics

Default Projects Folder

Base directory for quick-creating project folders

Data Folder

Where Yardmaster stores your settings, sessions, and database

Global Hotkey

System-wide shortcut to bring Yardmaster to the foreground

Quick Launch Hotkey

Floating composer hotkey for starting a new session from anywhere

Enable Quick Launch

Toggle the floating Quick Launch composer on or off

Remember unsent Quick Launch text

Keep your Quick Launch draft when you close the popup without sending, instead of starting blank

Quick Launch Default Project

Which project Quick Launch starts a new session in by default

Default project for quick tasks

Which Tasks project the Quick Launch Task popup files a quick capture into by default

Quick Launch tabs

Pick which tabs appear in the Quick Launch composer, reorder pinned tabs, and assign a global hotkey that opens a specific tab

Enable Quick Email

Send a short email from the Control Space composer in two keystrokes, pick a saved recipient, type a body, send

Quick Email Recipients

Up to 6 saved addresses you can reach from the Quick Email composer

Quick Email Recipient Hotkeys

Global per-recipient hotkeys open Quick Email pre-filled to that person

Quick Email Undo Window

After hitting send, Quick Email shows an Undo toast for this long before actually sending

Quick Email Show Subject by Default

When on, the Quick Email composer always shows a Subject field. When off, the body alone is enough and Subject is hidden behind a toggle

Quick Email Signature

A short sign-off appended to the bottom of every quick email you send, after a blank line

Data Retention

How long to keep data from finished sessions before automatically deleting it

Claude CLI Location

Where to find the Claude Code program on your computer. Usually auto-detected

Crash evidence recovery (always on)

Crash evidence left behind by a previous launch is folded into an on-device crash log at the next start. Nothing is emailed, and crash dumps never leave your computer

Launch on Startup

Automatically start when you log in to your computer

Reduce Terminal Popups

Stop stray terminal windows from popping up and stealing focus while agents run commands, switches the Windows default terminal to the classic Console Host (Windows only, reversible)

Auto-triage incoming bug reports

Email subject must use [BUG: slug] or [FR: slug] to be auto-routed into the bug intake project

Auto-Delete Old Scratchpad Notes

Permanently delete Quick Capture notes you haven't edited in a while. Off by default.

Delete notes I haven't edited in (days)

Notes whose last edit is older than this many days are permanently deleted on the next app startup and once a day after. Editing a note resets its clock.

Auto-Delete Old Message Content

Permanently erase old message content (SMS, email, chat, webhook, voice, and meeting recordings and transcripts) after a one-time notice and a 30-day grace. Off by default.

Delete message content older than (days)

Message content older than this many days is permanently erased on the daily cleanup, after the one-time notice and 30-day grace. There is no undo.

Enable bookmarks

Show the bookmarks toolbar button + popover. Off by default. When off, the bookmark icon is hidden and the popover is unreachable. Saved bookmark rows are preserved.

Toolbar

Choose which buttons are pinned to the header toolbar. Unpinned items are available in the overflow menu.

Keyboard Shortcuts

Customize key bindings for 50+ actions, including the one-handed (left-hand) layout preset

One-handed layout

Apply a left-hand-only keyboard layout to drive the thread without leaving the home row

Send on Caps Lock

Optional one-handed send key, press Caps Lock to send the message, just like Enter

Gmail Keyboard Shortcuts

Customize keyboard shortcuts for Gmail actions like archive, reply, compose, star, and navigation

Diff Review Keyboard Shortcuts

Customize keyboard shortcuts for code diff review actions like accept, reject, navigate files

OS-level hotkeys

Global, New Session, Quick Launch, and Jarvis hotkeys, fire even when Yardmaster is not focused

KMS Shortcuts

Customize keyboard shortcuts that fire inside the KMS notes panel (Bold, Italic, Insert link, Command palette, etc.). Defaults coexist with Yardmaster global shortcuts via panel-scoped dispatch.

KMS Panel Shortcuts

Rebind the KMS notes command-palette keys, Bold, Italic, Insert link, Command palette, search vault, find/replace.

KMS Editor Formatting Shortcuts

Rebind the KMS rich-text editor formatting keys, bulleted / numbered / task list, strikethrough, clear formatting, code block, blockquote, duplicate block.

Tasks Shortcuts

Customize the Tasks command-mode keys, snooze, deadline, archive, done, context, launch agent, break down, command line, focus moves, and the cheat-sheet.

Wake word detection

Say “Hey Jarvis” or “Computer” to activate voice commands hands-free

Read AI Messages Aloud

Have AI responses read aloud in a natural voice

Voice service

Choose which service reads messages aloud (Fish Audio, Grok, or ElevenLabs)

Fish Audio API Key

API key for the Fish Audio voice service

xAI API Key

API key for the Grok voice service

TTS Voice

Choose a voice model for text-to-speech playback

Grok Voice

Choose a preset Grok voice for text-to-speech playback

ElevenLabs Voice ID

ElevenLabs voice ID for text-to-speech playback

TTS Volume

Volume level for text-to-speech playback

TTS Speed

Playback speed for text-to-speech audio

Max characters

Limit how much of a message is read aloud (longer messages cost more)

Enable Voice Input

Speak to dictate messages or run hands-free voice commands

Speech Recognition Service

Service that converts speech into text (Deepgram, ElevenLabs, Groq Whisper, or the free Built-in engine)

How it listens

Choose between the free built-in option (the default, no account) and Picovoice (needs a paid access key)

After the Wake Word

Dictate the transcript into the active session, or run it as an in-app command

Voice AI Fallback

The AI that understands trickier voice commands

Voice Language

Language for speech recognition

How to start listening

Three ways to trigger voice input: wake word, toolbar microphone, or the Alt+V hotkey

Dictation delivery

Whether dictated speech is sent immediately or placed in the composer to review

Enable Jarvis Briefing

A spoken summary of a single session

Jarvis Briefing Model

Which Claude model summarizes the session. Haiku is faster; Sonnet is more nuanced.

Auto-play Jarvis briefing on session open

Speak a briefing when opening a session with new messages

Global hotkey

Whether the Jarvis briefing hotkey works when Yardmaster is unfocused

Key combination

Key combination that triggers the Jarvis briefing for the active session

Daily voice conversation cost cap

Daily spending limit for voice conversation responses, in dollars

Ask about a thread (voice)

Ask Yardmaster about a session by voice: hear a summary or ask a question, hands-free.

Ask the live agent (voice)

Queue a spoken question to a running session and get notified when its agent answers.

Enable Voice Report Back

Speak the AI response aloud after each voice command result

Voice Report Back max characters

Limit how many characters of the AI response are spoken

Voice Report Back daily cost cap

Daily spending limit for reading responses aloud, in dollars

Calibrate microphone

Tune how easily talking interrupts the spoken reply, matched to your mic and room.

Auto-read on first view

Automatically read the response when you first open a session that needs attention

Summarization mode

How long responses are read aloud: auto, always in full, or always summarized

Summary threshold

Responses shorter than this character count are read verbatim; longer responses are summarized

Confirmation speed

Playback speed for voice command confirmations

Spoken Narration auto-play

Automatically play the spoken recap when a new reply finishes in the open session

Spoken Narration prompt override

Override the built-in prompt used to write the spoken script for each reply

Push to Command

Hold a key anywhere on your computer, speak a command, release to run it

Push to Dictate

Hold a key anywhere on your computer, speak, release to type it into the session input

Default Assertiveness

How alarms surface when they fire.

Enable Alarms

Master switch for the Alarms feature

Default Alarm Sound

Sound played when an alarm fires

Default Snooze

How long to snooze a firing alarm by default

Bring Yardmaster to Foreground

Raise the Yardmaster window above other apps when an alarm fires

Pierce Focus Mode

Fire alarms immediately even during Focus Mode batching

Max Ringing Time

Auto-dismiss alarms to inbox after this many minutes

NLP Daily Cost Cap

Max daily spend on Haiku-assisted natural-language alarm parsing

Default Sound Play Mode

Play the alarm sound once or loop until you dismiss

Default Sound Duration

How long the alarm sound plays before stopping (or forever)

Default Fade-In / Crescendo

Ramp the alarm volume from silent to full over this many seconds, gentle wake or crescendo

Speak Alarm Label

Have the alarm read its label aloud when it fires (text-to-speech)

Snooze Cap

Maximum number of times an alarm can be snoozed before only Dismiss remains

Pomodoro Auto-Start Schedule

Allow weekday rules to auto-start Pomodoro runs at their scheduled times (turn off to silence the scheduler without deleting rules)

Pomodoro Phase-Change Foregrounding

Raise the Yardmaster window above other apps on each Pomodoro phase transition (focus → break and back). Off by default, most users prefer just the chime.

Auto-Enable Focus Mode During Focus Blocks

When on, starting a focus block silences notifications via Focus Mode; breaks release them. Manually toggling the bell icon during a run stops Pomodoro from managing it for the rest of the run. Per-preset overrides still win.

In-App Notifications

Choose which pop-up messages are shown in the app

Notification Behavior

When to show system notifications: always, backgrounded, or never

Notification Sound

Enable sounds and choose per-type notification tones

Sound Volume

Notification sound volume level

Usage Forecast

Predicted time until your Claude rate-limit window is exhausted

5-hour Exhaustion Warning

Notify me when the 5-hour bucket will exhaust within N hours (0 = off, max 5)

Weekly (7-day) Exhaustion Warning

Notify me when the weekly limit will exhaust within N hours (0 = off, max 168)

Cron failure alerts

Show a system notification and inbox card when a cron job fails permanently

Always fire cron failure alerts

Bypass Focus Mode batching for cron-failure alerts

Mobile notification chime

Play a short chime on mobile and web clients when a session needs attention

Push notifications

Receive OS notifications on mobile when sessions finish or need attention, even when the browser tab is in the background

Record-break notifications

Toast when you break a Stats record

Record-break inbox alerts

Add an inbox card when you set a new milestone Stats record

Limit repeat problem alerts

After you dismiss a problem or warning alert, keep the same one from coming back for 24 hours. Messages like team chat are never affected, and urgent alerts still get through.

Low-memory warnings

Warn me with an inbox card when this computer is running low on memory (at most once a day).

Incoming SMS notifications

Show an OS notification and play a chime when a text message arrives. Turn off to quiet SMS without silencing your other notifications.

AI cost alerts

Warn me with an inbox card when my AI spend for the day is unusually high (at most once a day). A genuine runaway always shows regardless.

Daily spend summary

Each morning, add an inbox card summarizing yesterday's AI spend, real API-key spend vs plan-included usage, broken down by provider. Off by default.

Stuck-install cleanup notices

Tell me with an inbox card when leftover dependency-install jobs from stopped sessions are cleared (the cleanup keeps running either way)

Screen-grab block notices

Tell me with an inbox card when an agent was blocked from bringing the app to the front while I was working in another app (the protection stays either way)

Anthropic outage alerts

Watch Anthropic's status page and warn me (inbox card + notification) when an Anthropic incident is making Claude sessions fail

New model alerts

Watch each provider you have configured for newly-released models and raise an inbox card with a one-click add-it-for-me action. With the live model catalog on, new models from every provider are added for you automatically instead

Kimi balance monitor

Watch your Moonshot (Kimi) API balance and raise an inbox card with a one-click Recharge button before it runs out

Kimi balance alert threshold

Get the Kimi low-balance alert once your Moonshot balance falls under this dollar amount

Show Alerts in sidebar

Show the Alerts project row in the left sidebar (off by default)

Active alerts

View the alerts your agents have raised

Recent notifications

See what just popped up and mute a category on the spot; resets when Yardmaster restarts

Google Calendar reminders

Get notified before an event starts, and when events are added or changed on your calendar

Calendar notification delivery

Deliver calendar notifications as an inbox alert, a toast pop-up, or both

Calendar reminder lead time

How many minutes before an event starts to send the "starting soon" reminder

Calendar added / changed event notices

Notify me when an event is added or changed on my calendar

Focus Mode

Batches inbox alerts so they fire on a count or time threshold instead of interrupting on every Needs-You session

Enable Focus Mode feature

Master kill-switch for Focus Mode. When off the toolbar pill is hidden and alerts always fire individually

Count threshold (sessions)

Number of Needs-You sessions that must accumulate before Focus Mode releases a batched alert

Time threshold (minutes)

Maximum minutes to wait before Focus Mode releases a batched alert even below the count threshold

Operator

Combine the count and time thresholds with AND (both must hit) or OR (either fires the batched alert)

Enable Doc Token Alerts

Master switch for warnings when always-loaded agent docs (CLAUDE.md, AGENTS.md, .claude/memory/MEMORY.md) cross a token threshold

Global threshold (tokens)

Global default size at which an always-loaded-docs alert fires. Defaults to 40,000 tokens

Per-project overrides

Override the global doc-token threshold for a single project that legitimately needs more headroom

Re-alert growth %

After you dismiss, only re-alert when the docs grow this percent past the size you dismissed at. Default 50%

Re-alert cadence (days)

Also re-alert this many days after dismissing if the docs are still over budget. 0 turns off the periodic reminder. Default 30 days

Enable MCP Overhead Alerts

Master switch for warnings when your MCP servers add up to too much overhead, estimated tool-definition tokens, active-session RAM, or unused-server count

Estimated token limit

Estimated tool-definition token overhead at which an MCP overhead alert fires. The figure is an estimate. 0 disables this check

Active-RAM limit (MB)

Active-session MCP process memory (in megabytes) at which an MCP overhead alert fires. Best-effort, skipped when no fresh sample exists. 0 disables this check

Unused-server limit

Number of configured-but-unused MCP servers (no tool calls in the usage window) at which an MCP overhead alert fires. 0 disables this check

Re-alert growth %

After you dismiss an MCP overhead alert, only re-alert when a tripping metric grows this percent past the value you dismissed at. Default 50%

Re-alert cadence (days)

Also re-alert this many days after dismissing an MCP overhead alert, if still over a limit. 0 turns off the periodic reminder. Default 30 days

Enable Skill Overhead Alerts

Master switch for warnings when your installed skills add up to too much overhead: estimated always-on frontmatter tokens, unused-skill count, or broken skills

Always-on token limit

Estimated always-on frontmatter-token overhead across all installed skills at which a skill overhead alert fires. The figure is an estimate. 0 disables this check

Unused-skill limit

Number of installed-but-unused skills (no invocation in the last 30 days) at which a skill overhead alert fires. 0 disables this check

Alert on broken skills

Alert when even one installed skill is broken: an unreadable header, no description, or a referenced file that is missing. On by default

Re-alert growth %

After you dismiss a skill overhead alert, only re-alert when a tripping metric grows this percent past the value you dismissed at. Default 50%

Re-alert cadence (days)

Also re-alert this many days after dismissing a skill overhead alert, if still over a limit. 0 turns off the periodic reminder. Default 30 days

Re-run Setup Wizard

Walk through account, toolchain, theme, and mobile access setup again

App Tours

Guided tours: Welcome, Sessions & Productivity, Voice Commands

Interactive Demo

See the dashboard in action with sample projects and sessions

Keyboard Demo

Walk the demo by pressing the app's real shortcuts, hands-on keyboard practice

KMS Quick Reference

Learn what KMS is, how agents interact with your vault, and set up your vault.

Login Accounts

Sign in with Anthropic OAuth to use your Claude plan

API Keys

Add Anthropic API keys for direct access

Anthropic API Key

API key required for AI features like titles, suggestions, spam filtering, automations

Usage & Rate Limits

View token usage, rate limits, and plan quotas per account

Extra Usage Alerts

Toast notification when extra usage credits are being consumed

Auto-Switch Account on Rate Limit

Switch to another account when rate-limited instead of waiting

Distribute Sessions Across Accounts

Spread new sessions across login accounts by usage instead of filling one first

Default Model by Account Plan

Start new sessions on a model chosen from the account plan: Max on Opus, Pro on Sonnet

Allow API Keys to Run Sessions

Control whether API key accounts can spawn Claude Code sessions

Show alternative AI providers

Master toggle for Codex, Gemini, Anti-Gravity, DeepSeek, Kimi, OpenCode, and Cursor in the new-session menu

Always use my default

Hide the tool/model/thinking picker on new sessions so they start on your default; you can still change one session before its first message

Tools in the picker

Choose which tools appear in the new-session picker (Claude Code is always available)

Gemini API Key

API key for Gemini CLI sessions

Allow Gemini sessions in any project

Opt-in toggle to allow spawning Gemini-provider sessions across all projects

Show Gemini virtual project

Adds a dedicated Gemini virtual project to the sidebar

Gemini CLI binary

Detect the Gemini CLI on your PATH (npm install -g @google/gemini-cli)

OpenAI API key

API key used for Codex CLI sessions

Allow Codex sessions in any project

Enable spawning Codex-provider sessions across all projects

Show Codex virtual project

Adds a dedicated Codex virtual project to the sidebar

Codex permission level

Choose how much access a Codex session gets and when it asks you: ask before every command, ask only when it reaches outside the project, or allow all

Codex CLI binary

Install or detect the Codex CLI

Anti-Gravity CLI binary

Detect the Anti-Gravity CLI and launch Google sign-in

Allow Anti-Gravity sessions in any project

Enable spawning Anti-Gravity-provider sessions across all projects

Show Anti-Gravity virtual project

Adds a dedicated Anti-Gravity virtual project to the sidebar

Give Codex and Anti-Gravity my Claude rules

Send Codex and Anti-Gravity sessions your personal CLAUDE.md and House Rules as well as the project's rules

DeepSeek API Key

API key for DeepSeek (Anthropic-compatible API) sessions

Allow DeepSeek sessions in any project

Opt-in toggle to expose DeepSeek as a session provider in the new-session menu

Kimi API Key

API key for Kimi (Moonshot Anthropic-compatible API) sessions

Allow Kimi sessions in any project

Opt-in toggle to expose Kimi (Moonshot) as a session provider in the new-session menu

GLM API Key

API key for GLM (Zhipu z.ai Anthropic-compatible API) sessions

Allow GLM sessions in any project

Opt-in toggle to expose GLM (Zhipu z.ai) as a session provider in the new-session menu

MiniMax API Key

API key for MiniMax (Anthropic-compatible API) sessions

Allow MiniMax sessions in any project

Opt-in toggle to expose MiniMax as a session provider in the new-session menu

Set up OpenCode for me

One click installs the OpenCode CLI, turns it on, and reuses your saved Anthropic key

OpenCode CLI binary

Detect the OpenCode CLI on your PATH

OpenCode model

Which model OpenCode runs: Claude (default), Kimi, GPT, or a custom provider/model

OpenRouter API Key (for other models)

OpenRouter key used when an openrouter/… OpenCode model is selected (Kimi, GPT, …)

OpenCode API Key (Anthropic)

Anthropic API key used by OpenCode CLI sessions (OpenCode runs Claude models)

Allow OpenCode sessions in any project

Opt-in toggle to expose OpenCode as a session provider in the new-session menu

Cursor CLI binary

Detect the cursor-agent CLI on your PATH

Cursor API Key

Cursor API key (from the Cursor dashboard) used by cursor-agent CLI sessions

Allow Cursor sessions in any project

Opt-in toggle to expose Cursor as a session provider in the new-session menu

Grok CLI binary

Install the grok CLI (xAI Grok Build) from x.ai/cli

xAI API Key (Grok)

xAI API key used by the grok CLI for background Grok Build sessions

Allow Grok sessions in any project

Opt-in toggle to expose Grok Build as a session provider in the new-session menu

Allow Hermes sessions in any project

Opt-in toggle to expose Hermes as a session provider in the new-session menu

Hermes CLI binary

Detect the Hermes CLI on your PATH (run hermes setup to pick a model and key)

Set up Pi for me

One click installs the Pi CLI, turns it on, and reuses your saved Anthropic key

Allow Pi sessions in any project

Opt-in toggle to expose Pi as a session provider in the new-session menu

Pi CLI binary

Detect the pi CLI on your PATH (npm i -g @earendil-works/pi-coding-agent)

Pi model

Which model Pi runs as a coding agent, picked from your configured provider

Identity

Your signed-in account: email, display name, role, and plan tier

Sign in with Google

Sign in to your account with Google from the signed-out Identity card

Sign in with GitHub

Sign in to your account with GitHub from the signed-out Identity card, when GitHub OAuth is configured

Edit profile

Change your display name and profile photo

Linked providers

Connect Google or GitHub to your account for single sign-on

Sign out

Sign out of your account

Delete my account

Permanently delete your account and all of your data across your devices. This cannot be undone

Firebase configuration

Paste the Firebase web config (API key, project id, OAuth client ids) to turn on the sign-in screen for this computer

Claude and Codex account tabs

Switch between the Claude and Codex account panels inside the Accounts section

Managed Codex accounts

Yardmaster-managed Codex (ChatGPT) accounts, each in its own isolated home, so you can keep several signed in and pick one per session

Managed Codex accounts (none yet)

Shown when you have no Yardmaster-managed Codex accounts yet, log in to add your first

Add a Codex account

Sign in to ChatGPT to add an Yardmaster-managed Codex account in its own isolated home

OpenAI API key (managed Codex)

Optional OpenAI API key that overrides your managed Codex (ChatGPT) logins for Codex sessions

Use the classic Accounts layout

Switch the Accounts page back to the previous flat layout instead of the grouped "Other AI tools" list

OpenRouter project spending limits

Review local lifetime managed-spending limits across billing connections in a project.

OpenRouter connections

Manage named encrypted OpenRouter connections and local managed spending limits in development builds.

OpenRouter model catalog

Browse, compare and favorite OpenRouter models and endpoint prices.

Your OpenRouter key

Paste your own OpenRouter key to power the app's built-in AI helpers (session titles, suggestions, reply drafts, daily digest).

Coaching Style

Set the overall tone, warm, neutral, or direct, for all coaching interviews

Daily Journal nudge time

Time of day your daily check-in and weekly review nudge appears

Weekly review day

Which day of the week your longer weekly review nudge appears

Coaching

Contextual suggestions to help you discover features and work more efficiently

AI Coach

Umbrella section grouping the coaching surfaces: Workflow Coach, AI Coaching, and the legacy Coaching nudges.

Enable Plain Speak

Master on/off switch, turning off hides everything beneath.

Show Plain Speak first

Open agent messages on the Plain Speak card instead of the original message by default.

Skip Plain Speak for phrases

Phrases that suppress the Plain Speak card when an agent message contains one (e.g. [PR-MERGE for PR-merge cards).

Plain Speak

Plain Speak, the agent-message overlay that rewrites turns into plain English.

Invite User

Send an invite to add a new user to your Yardmaster organization. Assign a role (owner, admin, member, trial) and tier.

Access Requests

Review and approve or deny pending access requests from users who want to join your organization.

User Roles & Tiers

Manage user roles (owner, admin, member, trial) and subscription tiers (enterprise, pro, starter, free) for your organization.

Auto-Continue Mid-Task

Automatically nudges the agent to keep going when it pauses mid-task without a question. Prevents unnecessary inbox interruptions during long jobs.

Auto-Recover When Out of Context

When a session fills its context window and stops mid-task without finishing, Yardmaster automatically frees space and continues it, so you see a finished reply instead of a blank.

Check agent replies

Pre-send fact check: a second model checks the claims in an agent reply against what the session did. Watch only shows what would have been sent back; Send back returns unbacked claims to the agent once.

Fact check route

Run fact checks on your Claude subscription (counts toward plan usage) or on your own Anthropic API key (billed per check).

Fact check model

The model that checks each reply: Sonnet (default), Haiku or Opus.

Daily check limit

The most fact checks run in one day (default 150).

Pause when plan usage is at or above

Skip fact checks while your Claude plan usage is at or above this percentage (default 80%).

Session Naming

How new sessions are named: AI-generated, first message, or numbered

Submit Key

Enter to send or Ctrl+Enter to send

Team Chat submit key

In Team Chat: Enter to send or Ctrl+Enter to send

Auto-continue lists

Continue bullet or numbered lists on a new line automatically

Fast typing (isolated composer)

Isolate the message box so typing stays smooth in large sessions

Composer buttons

Choose which action buttons show in the message box: attach files, quick replies and Decision mode. On a phone they collapse behind a single "+" menu, so hiding ones you never use frees space for the text box.

Clipboard history (Win+V)

Press Win+V to open your clipboard history anywhere. Show the clipboard icon in the composer, auto-paste the item you pick into the app you were in, and optionally keep your clipboard history across restarts (saved encrypted on this PC). In-development (Lab); Windows only.

Export to Word

Show an "Export to Word" option in the export menus (session ⋯ Exports submenu + the Export… button on docs, scratchpads, and messages). Markdown and PDF stay available. Off by default.

Mobile session arrows

Show prev/next session arrow buttons in the mobile session header

Default Model

Which Claude model to use for new sessions

Default Provider

Which AI engine new sessions start on, across all projects

Thinking Level

How much reasoning Claude uses before responding

Per-model thinking level

Set a default reasoning level for each model

Max Live Sessions Per Project

Maximum number of concurrent live sessions allowed per project

Stall Detection Time

How long without output before marking a session as stalled

Message History Limit

How many past messages to show when you open a session

Detect waiting agents

Suppress the needs-you inbox flip when the agent says it is waiting on something (build, tests, notification) without using the ScheduleWakeup tool

Recognize all waiting phrasings

Match the full range of waiting phrasings (catches more, occasional misfires) or only the official "Waiting on a background process" sentinel, official text only, near-zero misfires

AI second-opinion on waiting detection

When the regex waiting-detector is unsure, a cheap model (Haiku) double-checks before flipping a session to Needs-You and can catch waits the rules miss. Requires an AI key; off by default

Warn when a Chat Gets Large

Show a banner once a session context passes a size threshold

Context Warning Size (Thousands of Tokens)

The context size at which the large-chat warning banner first shows

Hold Automatic Resumes of Large Idle Chats

Hold cold automatic resumes instead of re-sending the whole conversation

Cold Resume Hold Size (Thousands of Tokens)

The context size above which a cold automatic resume is held

Budget Caps

Per-session cost limits: warning and stop thresholds in USD

Daily Limit (per account)

Per-account daily spend limit for pay-per-use API-key sessions (subscriptions exempt)

Daily Limit (per metered provider)

Per-provider daily spend limit for metered third-party providers (Kimi, DeepSeek, GLM, MiniMax, Meta)

Test Track daily limit

Daily limit on billed, pay-per-use spend from Test Track rounds (plan use is never counted)

Work on Separate Copies

Each session gets its own copy of the project files so multiple sessions cannot interfere with each other

Session Tags

Add short labels to sessions for quick categorisation. Tags appear as colored chips on the session header.

AI-Powered Search

Find sessions by meaning using local AI embeddings, not just keywords

Snooze notes

Show an optional note field in the snooze palette so you can leave a reminder for future-you.

Custom snooze shortcuts

Define your own snooze shorthand: a regex pattern that expands to a phrase the snooze box understands (e.g. nxwk8 becomes next Monday 8am).

Hide asides by default

Collapse side-question bubbles behind a one-line divider in the transcript by default.

Hide background check-ins

Collapse the short status message a session posts after its own background job finishes (e.g. "Tests green (exit 0). Waiting on lint.") to a one-line marker you can expand. Nothing is deleted; "Show system messages" reveals them. Off by default.

Hide repeated auto-waits

Collapse repeated "⏱ Auto-waited" messages (prose waits and scheduled wake-ups) to one-line markers you can expand, keeping the latest visible. The first wait (attached to your prompt) and the live/in-flight wait always stay visible. Nothing is deleted. Off by default.

Count repeated auto-waits

Fold the repeated "Auto-wait timed out" and "Wait timer reset" rows of one wait episode into a count on the ⏱ Auto-waited badge (e.g. "Auto-waited 2×") instead of adding a new row each time. Nothing is deleted. Off by default.

Max Consecutive Auto-Continues

Safety cap, forces inbox after this many consecutive auto-continues

Convert pasted rich text to Markdown

When pasting from Google Docs, Word, or web pages into the chat composer, convert the HTML formatting to Markdown. Off makes Ctrl+V behave like Ctrl+Shift+V.

Aside cost cap (USD)

Maximum cost per side question in USD. Asides exceeding this cap are aborted before spawn.

Show recipe pipeline lane sessions in sidebar

Reveal the per-lane sub-sessions spawned by recipe pipelines. Off by default to keep the sidebar clean, lanes auto-archive on clean completion. Notifications stay silent for lanes regardless.

Show X close button in session header

Show the X close button at the right of every session header. On by default. When off, close sessions with Ctrl+W or middle-click in the sidebar.

Show context indicator in session header

Show the context-percent ring in the session header so you can see how full the context window is at a glance. Hidden by default to keep the header clean.

Show voice button in session header

Show the voice/TTS speaker button in the session header for reading agent messages aloud. Hidden by default. Text-to-speech must also be enabled for the button to do anything.

Show subagent activity

Show a chip in the session UI when the agent has dispatched subagents (Task tool fan-outs) that have been running for more than ~15 seconds, AND a collapsible per-result card in the transcript when each subagent finishes (errors auto-expand, success collapsed). Off by default; both the chip and the cards are hidden. The underlying tracking that keeps the stall watchdog patient during long fan-outs runs regardless of this setting.

Silence recipe step sessions on clean completion

Auto-archive mainline recipe step sessions and the orchestrator when they finish cleanly with no errors, questions, or trouble markers (STEP_FAILED:, LANE_FAILED:, PREFLIGHT_FAILED:, NEEDS_ATTENTION:, RECIPE_NEEDS_ATTENTION:). Off by default, flip on to declutter the inbox after long audit runs.

Recipe approval default timeout (days)

Default wait time before a recipe approval gate auto-resolves. Applies only when the step does not specify its own timeout. 7 days by default. Range 1-90.

Mobbin MCP (design research)

Enable Mobbin MCP integration for Design Overhaul recipes. Searches real-world app screens, UI patterns, and flows for design inspiration. Requires a Mobbin Pro account.

Recipe concurrency warning

Show a confirmation dialog when starting a recipe while another is already running, since they share Claude account capacity. The dialog offers an opt-out checkbox.

Show keyboard-shortcut hints when I click

Show a brief "(or press X)" hint near the cursor when you click a control that has a keyboard shortcut. On by default.

Hotkey training mode

When on, clicking a button that also has a keyboard shortcut is blocked the first time and shows you the key to press instead. Click the same control again right away and it goes through. Keyboard shortcuts always work, and this never applies on a touch screen. Off by default.

Shared chat engine for non-Claude sessions

Codex, Pi, Gemini and other non-Claude sessions use the same conversation engine as Claude so their chats display consistently. On by default; turn it off to go back to the older, separate behavior of each engine.

Waiting period (minutes)

How long Yardmaster keeps a waiting session running on its own before it flips to your inbox, both when the agent says it is waiting on something and when it dispatches a background task. A background command can hold a session longer: up to that command's own time limit plus 2 minutes.

Show scroll-target overlay (debug)

Diagnostic overlay that briefly highlights where Chat v3 decides to land each time you open or scroll a session. Off by default; turn it on here to show it. It draws on top and never changes how scrolling works.

Lock to portrait on mobile

Keep the installed app in portrait on your phone instead of auto-rotating to landscape (Android only)

Keep-waiting button duration (minutes)

How long one press of the inbox "Keep waiting" button re-arms a silent wait. Separate from the initial waiting period

Auto check-in on waiting sessions

When a waiting session times out, automatically check in with the agent to re-check instead of flipping it straight to your inbox. Bounded: surfaces to you after 3 check-ins

Also check in after it reaches your inbox

When a waiting session has already surfaced to your inbox and sat untouched for a multiple of the waiting period, automatically check in with the agent. Bounded: surfaces to you after 3 check-ins

Check in after (multiple of the waiting period)

How long a surfaced session sits untouched in your inbox before Yardmaster auto-checks-in, as a multiple of the waiting period (e.g. 2× a 30-minute wait = 60 minutes)

Route Claude through the shared chat engine (advanced)

Advanced safety switch for an in-progress migration: routes your Claude sessions through the same shared chat engine the other providers use. Off by default, Claude keeps its battle-tested path. Reverts instantly with no restart.

Playwright MCP (browser automation)

Enable the built-in Playwright MCP server so sessions can automate a browser, navigate, click, fill forms, take snapshots, and run end-to-end checks.

WebApp Verify MCP (functional testing)

Built-in MCP server for automated web-app functional testing via Obscura. Verifies that a deployed or local app works correctly.

Session bar

Pin the session commands you use most (Snooze, Pause/Unpause, Send Later, Copy link, Aside Mode, Archive/Unarchive) to the session header bar next to the ⋯ button, moving them out of the overflow menu. Shown on desktop; on mobile they stay in the menu.

Match project color

Automatically color sessions to match their project color

Auto-Color Keyword Rules

Color sessions by keyword in their title

Catch-Up Card

Show a 4-line catch-up card pinned at the top of long sessions.

Catch-Up Card model

Override the LLM that generates the Catch-Up Card.

Show agent summary

Toggle the Did / Asks rows on the Catch-Up Card panel, what the agent did and what it is waiting on. Originally and Last asked always show. When off, the agent AI call is skipped.

Recent turns visible

How many back-and-forth turns to render on the Catch-Up Card panel. 1 = current only, 3 = current + 2 prior pairs. Range 1-5.

Originally prompt template

AI prompt that summarizes the user's first message into a 10-30 word plain-English anchor

Originally fallback length cap

Used only when the AI summary call fails. Caps the verbatim slice of the user's first message that gets stored as the fallback anchor for the Catch-Up Card.

Last asked prompt template

AI prompt that summarizes the user's most recent message

Agent prompt template

AI prompt that summarizes agent activity (did/asks)

Custom MCP Servers

Add, edit, or remove your own MCP servers. Custom servers appear in the per-project and per-session MCP picker alongside built-ins.

Back-test against a session

Run the continuous summary pipeline against an existing session to preview what it would produce.

Auto-focus on Needs You

Bring the app to the foreground when a session needs your attention.

Auto-Switch to Inbox

Automatically navigate to the inbox when another session needs attention, but only if you've been idle for at least 5 seconds. Won't interrupt scrolling, clicking, or typing

Auto-Focus Input on Session Switch

Automatically place cursor in the text box when switching sessions. When off, press R to focus the input (Gmail-style).

Agent Status Board

Show a live board of every agent Yardmaster has launched, its workspace, status, dev-pipeline phase, and to-do checklist, each with one click to jump to that session. Adds an Agent Board icon to the toolbar.

Auto-Resume Sessions on Restart

Automatically restart sessions that were active when the app was closed or crashed. Running sessions receive a continue nudge. Sessions waiting for your input are restored.

Keep Relaunching a Session Until It’s Back

When a session that was running fails to launch because the machine is momentarily overloaded (a transient OS hiccup), keep retrying, with a growing gap between tries, until it comes back, instead of giving up after a few attempts. Genuine errors still stop and turn red.

Agent Permission Level

How much a Claude agent can do on its own before Yardmaster stops to ask you. The built-in protection for sensitive files (SSH keys, credentials, system files) always applies, whatever you pick. Applies to Claude Code sessions.

Auto-Approve Plans

When Claude proposes a plan before writing code, approve it automatically instead of asking you first

Auto-Approve Config File Writes

Automatically approve when Claude wants to modify its own config files (.claude/, CLAUDE.md). When off, you'll be asked to approve each change

After Sending a Message

What happens after sending a message: stay, next session, or clear panel

Inbox Post-Send Behavior

After sending while in inbox: stay on session or auto-advance

Default Agent Instructions

Inject curated instruction presets into every Claude Code session, attribution removal, cloud safety, private repos, README standards.

Additional instructions for every session

Free-form custom text appended to every spawned session alongside preset instructions.

How copies are handled in git

Whether the generated agent-instruction copies are committed to git or kept out of it. Keeping them out avoids merge conflicts; committing them is for teammates who use other AI tools.

Tags

Curated library of session tags

Search tags

Filter the tag list

Sessions

Claude Code sessions needing attention

SMS

Incoming text messages via Pushbullet

Daily Digest

Daily briefing summaries

GitHub

GitHub notifications, PRs, and issues

Gmail

Gmail messages and threads

Cron Jobs

Scheduled cron job results

Automations

Automation rule results

CLI Pending

Pending CLI approval requests

Recipes

Recipe orchestration results

Question Widget

Show an interactive wizard for answering structured questions instead of typing responses manually

Format Question Options

Display lettered options (A, B, C) as clean indented sub-items instead of bullet points

Confirm before submitting with unanswered questions

Show a confirmation dialog when the question widget would submit a question you haven't answered.

Decision mode for new sessions

Agents explain each option of a decision with pros, cons and risk. You can switch it per session from the message box.

Keep question position stable

When navigating between multiple questions, lock the question text at a stable position so your eyes don't need to re-track. Pills extend below the locked area. Turn off to return to the legacy behavior where each question renders at its own height.

Auto-Detect Questions

Spot questions in the agent's reply automatically, even when the agent doesn't explicitly ask you.

QW miss reporting

Show the "Report missed question widget" menu item on agent messages. Captures parser misses as pending fixtures so the parser team can investigate.

QW format hint injection

Silently appends a short formatting hint to the FIRST message you send in each new session so the agent shapes its clarifying questions to trigger the QuestionWidget parser. The hint is invisible in your chat bubble, only the outgoing prompt carries it. Turn off if you don't want the hint reaching the model.

Highlight recommended option

Outline the option the agent recommends (its text ends with "(recommended)") with a soft green glow in the question widget. On by default.

Raised 3D buttons

Give the question widget's option buttons and Send button the same glossy, raised 3D look as the app's filled buttons, a soft top sheen and a drop shadow. On by default; turn off for flat buttons.

Quick Reply Button

Show a lightning bolt button next to Quick Replies that sends a message instantly.

Quick Reply Message

The message sent when you click the lightning bolt Quick Reply button in the session toolbar.

Number-key quick send

Show number badges in the Alt+S quick-reply picker and let you press 1-9 to send that reply instantly.

Confirm on a blank session

On a blank session, the lightning bolt quick reply asks "Are you sure?" first, click it again to send, so a stray click cannot fire it as the first message.

Channel Filters

Master toggle for automatic message processing with rules or AI

Filter Rules

Create rules to automatically process incoming messages

AI Filters

Use AI to classify messages and trigger actions automatically

Filter Activity Log

View run history and results for all filters

Forward to email

Forward SMS or messages to one or more email addresses via Gmail (v1: Gmail only, max 20 per hour)

Automation Credentials

Saved encrypted secrets used by Automation Engine v2 actions (Slack bot tokens, GitHub PATs, SMTP relays). Configure webhook URLs, username and private key fields, plus the allow high-risk credential kinds gate.

Cooldown between fires (seconds)

Minimum seconds between consecutive firings of the same automation rule, preventing rapid re-triggers.

File Viewer Word Wrap

Wrap long lines in the file viewer instead of horizontal scrolling

Copy Full File Paths

Copy absolute paths to clipboard instead of relative paths from agent output

File Viewer Minimap

Show the VSCode-style minimap overview on the right of the file viewer

Word Wrap

Wrap long lines instead of horizontal scrolling

Syntax Highlighting

Apply language-aware syntax coloring to diff content

Diff Minimap

Show the VSCode-style minimap overview on the modified side of the diff

Default View Mode

How diffs are displayed: unified or side-by-side

Diff Font Size

Default font size for diff content

Tab Display Width

How many spaces wide a tab character appears in diffs

Lite mode (optimize for a less powerful computer)

One switch to make Yardmaster lighter on low-spec machines: turns on Low Power Mode, holds new sessions when memory is low, reclaims RAM from idle sessions (Windows), staggers session loading, and stops keeping extra chats pre-built in the background. Turn it off to restore your previous settings. Startup-related changes take effect on the next launch.

Keep my computer smooth under load (recommended)

When many sessions run at once, Yardmaster eases up on its own background monitoring so your machine stays responsive, without changing anything your sessions produce. On by default; turn off for the older, always-on monitoring.

Speed up agent commands (skip the redundant safety check when it cannot apply)

When on, Yardmaster skips a slow per-command security check for commands that cannot possibly trigger it, so each agent command is much cheaper. It still runs the full check, unchanged, for anything git or master related. It manages a hook in your global Claude config automatically, with a backup and an instant off switch, and does nothing on machines that do not have that heavy check installed.

GPU Acceleration

Uses your GPU for hardware-accelerated rendering. Disabling this falls back to software rendering with a significant performance cost.

Low Power Mode

Disables GPU-expensive visual effects (background blur, heavy animations, decorative shadows) for smoother rendering on integrated graphics and low-power laptops. Takes effect immediately, no restart required.

Fast session switching

Makes clicking between sessions feel snappier by skipping redundant re-rendering of the session panels you are not switching to. On by default, if you ever want the previous behavior back, flip it off and it reverts instantly. No restart, and no effect on what your sessions actually do.

Per-session update isolation

Cuts the cost of each live update when many sessions stream at once, so a busy agent only wakes its own panel instead of nudging every open one. Off by default; turn it on to try it. It re-wires instantly with no restart, and it never changes what your sessions do.

Smooth streaming output (recommended)

Paints rapid streaming text at 30 frames per second instead of on every token, which makes fast Claude responses noticeably smoother and keeps the rest of the app responsive while many sessions stream at once. Default on. Turn off only if you want every keystroke painted as it arrives, costs more CPU and offers no correctness benefit.

Background session refresh (advanced)

How often text from sessions you are NOT currently looking at gets repainted, in milliseconds. Higher = less background work when many sessions stream at once; the session you are viewing always stays smooth. When you switch to a background session it catches up within this window. Default 500. Only applies when smooth streaming output is on.

Real-conversation layout

Show just your messages and the agent’s final replies. Per-turn tool work folds into a collapsed “N actions” pill you can expand on demand. The single biggest render-cost reduction on busy sessions. Off = the legacy stream of every intermediate step.

Keep recent sessions instant

Keep the sessions you’ve looked at recently fully built in memory (the way a browser keeps background tabs ready) so switching back to one is instant instead of rebuilding it. Only applies to sessions that aren’t actively streaming, and a small limit caps how many are held so memory stays bounded. On by default.

Keep The Vault loaded for instant switching

Keeps your Vault editor loaded in the background once you’ve opened it, so switching back into The Vault is an instant flip instead of rebuilding the editor every time. Holds one editor in memory, and only after you first open The Vault. Desktop only. On by default; turn off to load The Vault fresh each time (the old behavior).

Keep recently-opened panels loaded for instant switching

Keeps recently-opened panels, AI Coaching, Calendar, Drive, and Sheets, loaded in the background once you’ve opened them, so switching back into one is an instant flip instead of rebuilding and re-fetching it every time. Each panel is held in memory only after its first open. Desktop only. On by default; turn off to load each panel fresh every time (the old behavior).

Keep Tasks loaded for instant switching

Keeps your Tasks panel loaded in the background once you’ve opened it, so switching back into it is an instant flip instead of rebuilding the list every time. Desktop only. On by default; turn off to load Tasks fresh each time (the old behavior).

Show checkboxes on tasks

Shows the little checkbox at the start of every task. On by default; turn it off for a cleaner list with no checkboxes. You can still finish a task with the keyboard (x or e), a middle-click on the row, or the row menu.

Pre-render inbox sessions

Pre-builds your most recent inbox (needs-you) sessions in the background so switching into them is instant. A built-in limit keeps the app fast even when the inbox is large, older inbox sessions load in about half a second when you open them. On by default; turn off to do no inbox pre-rendering at all.

Keep chat responsive under heavy load

When you have many sessions open at once, Yardmaster stops the ones you can't see from doing background work every time another session sends a message. This keeps scrolling smooth and the chat landing in the right place no matter how many sessions are running. On by default; turn off to mount every hidden session in full like before.

Pre-load new sessions

Keeps one blank session ready in each project so starting a new session (Ctrl+T) is instant. When off, new sessions spawn on demand, a bit slower but uses fewer resources. On by default.

Instant new session (mobile)

On your phone, tapping New Session opens an empty message box instantly, no waiting, and you can start as many as you like. The session itself is created the moment you send your first message, so an un-sent box never clutters your list. On by default; turn off to go back to preparing the session as soon as you tap.

Stagger session loading at startup

Spreads out the building of background session panels after the app starts or reloads, instead of building all of them at once (which can freeze the window for several seconds on launch). The session you’re looking at always loads first and immediately; the rest follow within a few seconds. Off by default while this is being evaluated, turn it on to try it, turn it off to go back to exactly the old behavior.

Reclaim RAM from idle sessions (Windows only)

Periodically asks Windows to evict the working set of Claude CLI child processes that have been idle for at least this many minutes. Reclaims physical RAM without killing the session, pages re-load on next use (~80 ms). Set to 0 to disable. No effect on macOS or Linux.

Release idle sessions to reclaim more RAM

Fully releases the background Claude engine of a session that finished its turn and has been waiting on you for at least this many minutes, reclaiming all of its RAM (~250-450 MB each), not just trimming it. The conversation stays on screen; your next message reloads it in the background (a slightly longer first reply). Sessions waiting on a question or plan keep their engine. Set to 0 to disable. Works on every platform.

Reclaim RAM automatically when memory runs low

When your computer's free memory drops and stays low, Yardmaster trims the memory its own idle sessions are holding, a safe, automatic tidy-up that never touches a session mid-reply. On by default. No effect on macOS or Linux.

Reclaim when free memory drops below

The trigger point. When free RAM stays below this percentage of your total memory, the automatic reclaim kicks in. Higher = reclaims sooner; lower = only under real pressure.

Cap runaway processes per session (Windows only)

Puts a kernel-level ceiling on how many processes a single Yardmaster session can run at once, so one misbehaving agent (a runaway loop, a fork bomb, an over-parallel command) can't flood the machine and freeze it. The ceiling is generous, far above any normal session, so it only ever stops a genuine runaway; new processes past it simply fail until some finish, and nothing already running is killed. No effect on macOS or Linux.

Cap CPU when many sessions are running (Windows only)

Soft-caps the combined CPU usage of all Yardmaster Claude CLI processes (sessions, subagents, test runs, builds, everything they spawn) when the host is contended. An isolated single session can still use the full machine; the cap only kicks in when other things want CPU. Below-normal process priority is also applied so Yardmaster yields to your foreground apps. No effect on macOS or Linux.

Max combined CPU %

Maximum percentage of total host CPU that all Yardmaster Claude processes can use combined when contended. Lower = friendlier to your foreground apps, slower for batch work. Range 25-90, step 5.

Strict ceiling, never exceed the cap, even on an idle machine

By default the cap is soft: it only kicks in when other apps want CPU, so a single session you are waiting on can still use the whole machine. Turn this on to make the cap a hard ceiling, Yardmaster sessions can never go above the percentage above, even when nothing else is running. Guarantees the limit, but can slow a lone session. No effect on macOS or Linux.

Boost Yardmaster interface priority (Windows only)

Raises Yardmaster's own processes (main + renderer + GPU + utility) to Above-Normal priority so they win scheduler fights against Claude CLI children at Below-Normal. Keeps the Yardmaster window, and other foreground apps like Chrome, snappy when many sessions are saturating the machine. No effect on macOS or Linux.

Keep Yardmaster resident in memory (Windows only)

Locks Yardmaster's database in physical RAM (Windows) so it can't be pushed out to the disk pagefile and then re-read from a busy disk, the cause of the occasional whole-app freeze under heavy multi-agent load. On by default: it auto-sizes to your machine, caps itself so it can't crowd out other apps, and turns itself off on low-memory PCs. Only Yardmaster's own engine is pinned, never your agents. No effect on macOS or Linux.

Serve the database from private memory (Windows only)

Serves Yardmaster's database from a large block of private memory instead of mapping it from the file on disk. Under heavy multi-agent disk load Windows evicts the file-mapped database from RAM and the app freezes while it re-reads from the busy disk, private memory can't be evicted that way, so the database stays resident and the freezes stop. Opt-in and fully reversible: turn it off and restart to go back. Takes effect on the next restart. No effect on macOS or Linux.

Run the database off the UI thread

Runs Yardmaster's database reads on a background thread instead of the one that draws the window. Leave this OFF: in real multi-session use it made the app much SLOWER, reads wait in a single line behind each other, and with "Dedicate a CPU core" on, the database work shares that one core with the window itself. It stays available only as an experiment for light workloads. Takes effect immediately, on or off, no restart needed.

A/B: force the off-thread DB worker ON (dev measurement)

Diagnostics measurement lever, shown only in dev builds. Forces the off-thread DB worker ON right now, the same as launching with AMC_ENABLE_DB_WORKER=1, bypassing the migration safety interlock and the RAM floor, so it can be A/B-measured from the app with no terminal. Applies live (no restart); for a clean measurement window, restart Yardmaster after toggling. The AMC_DISABLE_DB_WORKER kill switch still overrides this, and it is NOT the shipping default.

Let sessions yield to your foreground apps (recommended)

Runs every Claude session, and everything it spawns (searches, git, builds, tests, subagents), at Below-Normal priority, so it gives way to the Yardmaster window and your other apps when the machine is busy. A session you are waiting on alone still uses the full machine; it only steps aside when something else needs the CPU. This is the lighter, always-safe alternative to a hard CPU cap. Works on Windows and macOS (no effect on Linux).

Let sessions yield the disk to the app (recommended)

The disk-and-memory sibling of the priority setting above. Runs every Claude session, and everything it spawns, at low disk priority and below-normal memory priority, so when the app needs to read from disk (or pull its own memory back in) it goes first, and Windows frees the sessions’ cached files before the app’s. This is what keeps typing and clicking responsive while many sessions hammer the disk. Sessions at full speed when the app is idle; they only queue behind it under load. Windows only.

Pace heavy background jobs (recommended)

Limits how many heavy background jobs, test, typecheck, lint and build runs, plus cloud-upload preparation, can START at the same moment, letting them through in waves instead of all at once. When dozens fire together they spawn over a thousand short-lived helper processes in about a second, which can freeze the whole computer for a minute even though each job already runs at low priority. Your live sessions are never paced, only their background job runs are, and a paced job waits briefly instead of failing. Turning this off restores the old free-for-all.

Throttle agent searches (recommended)

When many sessions search your code at once (ripgrep / grep), Yardmaster lets them through a few at a time and at low priority so a swarm can't saturate your disk. On by default. Turn off to let your agents' searches run at full speed and unqueued, faster searches, but a big swarm can bog down this computer. Junk folders like node_modules are always skipped either way. Takes effect on your next new session.

Throttle agent git commands (recommended)

Caps how many git commands your sessions run at once so a burst can't hammer the disk and freeze this computer. On by default. Turn off to let your agents run git unrestricted, faster, but a big swarm can bog down this computer. Takes effect on your next new session.

Throttle agent installs (recommended)

Paces the npm / npx / pnpm installs your sessions start so a burst of them can't storm this computer. On by default. Turn off to let installs start immediately, faster, but a big swarm can bog down this computer. Takes effect on your next new session.

Throttle agent shell commands (recommended)

Briefly paces the shell commands your sessions run when this computer is already busy, so a command storm can't overwhelm it. On by default. Turn off to let every agent command run the instant it's issued, faster, but a big swarm can bog down this computer. Takes effect on your next new session.

Batch agent 'find' commands (recommended)

Rewrites your agents’ find commands to run their per-file work in efficient batches instead of launching one process per file, one runaway find can otherwise spawn tens of thousands of processes and freeze this computer. On by default and actually faster. Only turn off if a specific command genuinely needs the old one-process-per-file behavior. Takes effect on your next new session.

Pace new-session bursts (recommended)

When many sessions start at once (a recipe fan-out, a mass restart), Yardmaster starts them a few at a time so the burst can't freeze this computer. On by default. Turn off to start every new session immediately. A single session you open yourself is never delayed either way.

Pace worktree building (recommended)

When many isolated sessions start at once, Yardmaster builds their worktrees a few at a time so the burst can't saturate your disk and freeze this computer. On by default. Turn off to build them all at once, faster session starts, but a big burst can bog down this computer. Takes effect immediately.

Hold new sessions at the memory cliff (recommended)

A last-resort safety hold: when this computer is a hair away from running out of memory, Yardmaster briefly pauses starting a new session so the low-memory moment does not crash the whole app. It does nothing at all until you are genuinely at that cliff, so it has no everyday cost. Strongly recommended to leave ON, turning it off means a new session started at exactly that moment can crash Yardmaster. Takes effect immediately.

Speed up batch merges

While "merge all ready branches" runs, gives that session Above-Normal CPU priority and briefly drops your other sessions to Below-Normal so the batch finishes faster on a busy machine. Other sessions keep running, they just yield. Reverts automatically when the merge ends, after a 30-minute safety timeout, or if you turn this off. On macOS it lowers your other sessions but cannot raise the merge session above normal, so the speed-up is gentler; no effect on Linux.

Show the in-app worktree cleanup dashboard

Turns on the in-app Worktree Cleanup dashboard: a live view of the automatic cleanup that reclaims disk by retiring finished git worktrees, its run history, a grid explaining why each worktree is kept or reaped, one-click recovery of a mistakenly-removed worktree, and a "run cleanup now" button. The cleanup itself runs on a schedule regardless; this only adds the dashboard and controls. Windows only; off by default.

Show a notification when a worktree is cleaned up on archive

When you archive a session that already has a pull request, the worktree is cleaned up automatically. This shows a brief notification when that happens. Turn it off for fully silent cleanup.

Push sessions to efficient cores when CPU is high (Windows, hybrid CPUs)

On hybrid CPUs (Intel 12th-gen+ Alder Lake, Raptor Lake, Core Ultra), confines Claude CLI sessions to the efficiency (E) cores when host CPU averages above 85%, releasing them back to all cores when it drops below 60%. Lets sessions keep grinding while the performance (P) cores stay available for Yardmaster and Chrome. No effect on non-hybrid CPUs or non-Windows systems. The setting is safe to leave on; it stays a no-op when the host doesn't support it.

Reserve a CPU core for Yardmaster so the window never freezes (Windows, hybrid CPUs)

Sets aside a full performance (P) core just for Yardmaster's window, so a swarm of running sessions can never freeze the interface by crowding it off the CPU. Your sessions are kept off the reserved core and share the rest: they give up about 6% of the cores and keep running normally, while Yardmaster always has a clear core to stay responsive. Best for heavy multi-session use on a powerful hybrid CPU (Intel 12th-gen+ Alder Lake, Raptor Lake, Core Ultra). No effect on non-hybrid CPUs or non-Windows systems, and it steps aside automatically during a CPU burst. Off by default.

Make the reserved core truly dedicated: keep Yardmaster running on it (Windows, hybrid CPUs)

Goes one step further than reserving a core. Reserving keeps your sessions OFF the core, but doesn't guarantee Yardmaster actually uses it. Under a heavy swarm the window can still stutter while the spare core sits idle. This pins Yardmaster's own window onto the reserved core and KEEPS it there the whole time this is on, so Yardmaster always has a fast core kept clear of your sessions to stay responsive, not just drifting onto it sometimes. (Windows and other apps still run on that core too; no setting can fence those off. What this guarantees is your sessions stay off it and Yardmaster stays on it.) The trade-off: while on, Yardmaster sticks to that one performance core, so on an otherwise-idle machine a heavy redraw isn't spread across every core. Turn it off to let Yardmaster use them all. It's the piece the priority boost can't do on its own. Turning this on reserves at least one core automatically. Needs the reserved-core setting above and a hybrid Intel/AMD CPU. Off by default.

Ease RAM pressure by paging busy sessions to disk (Windows only)

When your machine runs low on memory, Yardmaster tells busy sessions (and the tests they run) to keep less in fast RAM and spill the rest to the disk pagefile. Sessions keep running; they just slow down until memory frees up. Kicks in only above ~85% RAM use and lifts automatically below ~70%. Never affects Yardmaster itself, and never cancels, blocks, or kills anything. No effect on macOS or Linux.

Hold new sessions when memory is low

When your computer is running low on memory, pause starting NEW sessions that arrive in a burst (auto-created bug/feature sessions, recipe runs, mass restarts) until memory frees up, so a flood of sessions can't drive the machine into a freeze. A single session you open by hand is never delayed, and any held session always starts within a few seconds. Sessions resuming after an app restart are always paced this way regardless of this toggle. Works on every platform.

Memory threshold %

Start holding new burst sessions once total system memory in use reaches this percentage. Lower = more cautious (holds sooner); higher = only holds when memory is nearly full. Range 50-95, step 5.

Spacing between sessions when many restart at once

When several sessions come back at the same time (restored after an app restart, auto-recovered from a rate limit, or restarted together from Manage sessions), this is how long to wait between starting each one, so a big batch does not storm your computer all at once. Sessions you start or restart by hand, one at a time, are never delayed. Set to 0 for no wait. Takes effect immediately, no restart needed.

Wait between automation-created new sessions

How long to wait between starting brand-new sessions that an agent or automation creates in a burst (for example a script spawning many sessions in a row). Keeps rapid automated spawns from piling up all at once. Sessions you open yourself are never affected. Set to 0 to start them immediately. Takes effect immediately, no restart needed.

Stop sessions from loading unused background tools

Each session normally starts every tool server listed in your global Claude config, including ones it never uses, like the Playwright browser tool, which can take roughly 95 MB of memory each. With many sessions open those idle tool servers pile up into gigabytes of wasted memory and can slow the whole computer down. When on, each session loads only the tools Yardmaster actually set up for it, so the waste can not accumulate (and sessions start a touch faster). Your Yardmaster tools (memory, The Vault, Google, and custom servers) keep working exactly as before. Applies to Claude sessions; Codex, Gemini, and Cursor are unaffected. On by default.

Keep computer awake while sessions are running

Prevents your computer from going to sleep while one or more sessions are running, so long jobs keep working when you step away. The display can still sleep to save power; only system sleep is blocked. Sleep returns to normal the moment every session finishes. Works on Windows and macOS; on Linux it depends on your desktop environment. Off by default.

Isolated session view (experimental)

Render the active session in its own background process inside the main window, so it stays responsive while the rest of the app is busy. Takes effect on the next session you click.

Check for silent sessions

Checks every few minutes whether a silent Claude session is still answering, and marks it Not responding if it stops. Turn off only if it misbehaves.

Share Brain offers

At the end of a turn, Share Brain reads the files the agent just wrote to see whether one is a finished page worth offering. It runs on this computer, one turn at a time, and is capped per turn. Turn it off to stop that work and every new offer at once. On by default; only shown while Share Brain is turned on in Lab.

Graphics acceleration auto-disabled

Banner shown after repeated startup crashes auto-disable GPU acceleration, lets you re-enable hardware acceleration once your graphics drivers are up to date.

Fast dev mode (disable React double-render)

Dev build only, never affects the production app or your users. Turns off React's StrictMode double-render so the app feels snappier while testing.

Sync agent instructions across tools

Mirror your canonical agent instructions file (CLAUDE.md or AGENTS.md) into the filenames other AI coding tools read, Codex, Gemini CLI, Cursor, Cline, Copilot, Windsurf. Off by default.

Enable Dev Pipeline skill

Install the bundled Dev Pipeline skill, a guided end-to-end dev workflow with five approval gates. Off by default.

Auto-approve every gate

Turn auto-approval on (or off) for all five Dev Pipeline gates at once, Plan, Red Team, Build, Elegance, Docs. Each gate also toggles independently; a gate that asks you a question is never auto-approved.

Remind every session to use the pipeline

Add a short note to every session telling it to follow the Dev Pipeline when doing software development, and to ignore it otherwise. Skips the in-app helper bots.

Add the pipeline quick replies

Add the "Red Team" and "Full Documentation Prompt" quick replies if they are missing. Turning this off leaves any you already have.

Add the workflow auto-replies

Set up a few automatic replies so a Dev Pipeline run keeps going without you typing the usual answers, they start and approve the red-team review, pick how the build runs, have each phase ask all its questions at once, and confirm the final merge-ready check. Skips any you already have; turning it off removes only the ones it added.

Save the pipeline note into the repo's files

Keep a short marked note in each enabled repo's CLAUDE.md (or AGENTS.md) telling agents to route work through the Dev Pipeline. Added when a session starts there and removed when you turn it off.

Auto-approve the plan gate

Auto-approve the Dev Pipeline Plan gate (GATE 1) on its own, so a run can start building without you approving the plan. Off by default; a plan that asks you a question is never auto-approved.

Always hide everything before the marker (even without tool activity)

A sub-option of "Show only the final message": when the final-answer marker is present, hide EVERYTHING before it, even on a plain answer with no tool steps, folding it into a collapsible row you can expand. Nothing is deleted; it stays one click away. Off by default.

Show every final message when an agent posts more than one

A sub-option of "Show only the final message": when an agent marks more than one final message in a single turn (say a report, then a follow-up), show ALL of them instead of only the last, with the work between them tucked into the collapsible row. Nothing is ever deleted. Off by default.

Show marker location (debug)

A sub-option of "Show only the final message": draw a distinct divider showing exactly where an agent placed the final-answer marker instead of hiding it. Nothing else changes, the same content stays folded, and it only appears on turns where the agent actually printed the marker. For debugging where agents put the marker. Off by default.

Enable Browser

Show the Browser entry in the sidebar, an embedded webview with a persistent cookie jar.

Running Apps

Show the Running Apps entry in the sidebar, a live list of the dev servers Yardmaster launched through the bundled portless proxy at their .localhost URLs.

Show Job Monitor in sidebar

Show the Job Monitor entry in the sidebar, a read-only dashboard of running test/build processes.

Show Scratchpads in sidebar

Show the Scratchpads entry in the sidebar. Scratchpads stays reachable from the toolbar button and its hotkey regardless.

Enable Drip

Surfaces the Drip virtual project: queue mixed content (text, links, files, folders) into named queues that release into the inbox on a user-chosen cadence. Turn off to hide it and stop every drip from releasing.

Render HTML files live

When on, a drip item that is an .html file renders live in a sandboxed preview, HTML, CSS and JavaScript run, but the page is isolated (no network, no access to your app). Off shows the file as a plain chip.

Enable Scheduled Messages

Surfaces the Scheduled Messages virtual project, compose a rich-markdown note to yourself and have it dropped into your inbox once or on a recurring schedule.

Allow agents to raise inbox alerts

Agents drop persistent inbox rows via POST /alert

Suggest saving repeated phrases as quick replies

When you type the same phrase many times, an inbox card offers to save it as a quick reply, prefilled and ready to approve.

Repeat threshold

How many times a phrase must repeat before the quick-reply nudge appears. Range 2-100; default 10.

Suggest removing quick replies you no longer use

When a saved quick reply goes unused for 30+ days, an inbox card offers to remove it (one tap, with undo).

Pause all drips

Master pause for every drip. When on, drips keep their cadence but skip every release until you turn this off.

Enable Google Workspace Tools

Gives sessions MCP tools for Google Drive, Docs, and Sheets

Surface stuck silent sessions

Manually clears the silent flag on every recipe session stuck in an attention status (needs_you / error / stalled) without waiting for the 6h watchdog. Makes them reappear in the sidebar and inbox.

Make sessions Yardmaster-aware

Adds a short note to every session so the agent knows it is running in Yardmaster and how to reach it.

Enable PR Merge Queue

Adds the PR Merge Queue sidebar entry and toolbar icon for triaging open pull requests across configured GitHub repos.

Enable Team Time

World clock dashboard for managing team members across timezones with live clocks, working-hours status, and a meeting planner.

Sticky Notes

Enable floating sticky notes on the desktop

Enable Tasks Outliner

Show the Tasks entry in the sidebar, nested checklist with due dates, snoozes, and a tasks.md mirror on disk.

Screen Capture options (moved into the tool)

Shortcuts, recording quality, frame rate, webcam shape, click ring, countdown, watermark, intro card, auto-zoom, click sound, auto-DND, export folder, link expiry, and the Vimeo token now live inside Screen Capture, open it and click the gear icon in its header.

Enable RepoGuard

Show the RepoGuard sidebar entry, polls Sentry for code-level alerts and surfaces them for triage by an agent.

Enable Bug Intake

Show the Bug Intake sidebar entry, collects bug reports from configured Sentry sources and triages them via spawned agents.

"Needs you" window (hours)

How many hours ahead the Tasks "Needs you" sidebar entry surfaces upcoming due tasks. Anything overdue is always included.

Enable Knowledge Management System (KMS)

Surfaces the KMS virtual project, a read-only view over a local Markdown vault with full-text search across all notes.

Enable Marketplace

Show the Marketplace sidebar entry and the marketplace browser under Settings > Plugins for discovering and installing community plugins. On by default. When off, installed plugins stay installed and keep running.

Check for plugin updates

Automatically check the marketplace for newer versions of your installed plugins on launch and periodically, then show a badge and a one-time notice. It never updates a plugin without your click.

Enable Marketplace Review

Show the Marketplace Review sidebar entry, the admin queue for approving or rejecting submitted plugins. Off by default.

Vault root folder

Absolute path to the Markdown vault folder KMS indexes. Must exist on disk; symlinks and system paths are rejected.

Other stored indexes

Folders KMS indexed before you changed the vault. They are not searched or watched, but still take up space in the database. Removing one never touches your files on disk.

Watch vault for changes

Re-index notes automatically as files are added, edited, or removed. When off, the vault is indexed once at startup and never re-scanned until next launch.

Only keep pinned notes as tabs

Opening a note previews it in a single tab that the next note replaces, so tabs stay tidy instead of piling up. Pin a note to keep it as a permanent tab.

KMS summary daily cost cap

Maximum USD spent per day generating KMS AI summaries on the active API-key account. When the daily total reaches this figure, summary generation pauses until local midnight. Spans haiku-nothari-summary and haiku-nothari-bulk sources.

KMS bulk-seed cost cap

Per-run cap (USD) on the one-time KMS bulk-seed worker that summarizes every note in the vault. Overrides the per-day summary cap for the duration of the run; the run halts with `cost-cap-hit` once the total reaches this figure.

Auto-regenerate stale summaries

When a KMS note drifts far enough from its stored summary, a background worker regenerates it on a schedule. Notes marked user-edited are never touched. Off by default, summaries otherwise refresh only on a manual Regenerate.

Auto-regen scan interval

How often (in minutes) the KMS auto-regen worker scans the vault for stale summaries. Stored internally as milliseconds. Applies only while auto-regenerate is enabled. Range 1-1440 minutes.

Auto-regen per-note cooldown

Minimum gap (in minutes) before the same KMS note can be auto-regenerated again, so one rapidly-edited note cannot dominate the worker budget. Stored internally as milliseconds. Applies only while auto-regenerate is enabled. Range 1-1440 minutes.

Expose Vault to spawned agents

Lets every spawned Claude CLI session read your KMS vault via the bundled MCP server. Adds five read-only tools (kms_search, kms_read, kms_list_tags, kms_list_recent, kms_get_backlinks). Hidden notes are excluded. No write access, destructive tools land in a later sub-PR behind a separate approval flow. Off by default.

Quick Reference Guide

Learn about KMS features and agent integration.

KMS appearance

Give the KMS its own look, a visual theme, light/dark mode, and editor reading options, independent of the rest of the app.

KMS visual theme

Pick the KMS visual theme, including imported custom themes, independent of the app theme.

KMS light / dark

Choose the KMS light or dark mode, or follow the app.

KMS editor font

Choose the KMS editor font, sans-serif, serif, or monospace.

KMS editor text size

Set the KMS editor text size, small, medium, large, or extra large.

KMS editor width

Set the KMS editor reading width, narrow, normal, wide, or full width.

KMS page background

Set the KMS editor page background, follow theme, paper, or sepia.

File tree side

Which side the KMS file list sits on, in the KMS panel and the pop-out window. Left by default.

Cross off completed checklist items

Checked KMS task-list items show a strikethrough and dimmed text; unchecking restores the original look. On by default.

AI Knowledge Digest

Periodically scans your sessions, extracts notable insights, and writes a curated Markdown digest into your vault.

Daily digest

Enable or disable the daily knowledge aggregation digest.

Daily digest hour

What hour to run the daily knowledge aggregation digest.

Weekly digest

Enable or disable the weekly knowledge aggregation digest.

Weekly digest schedule

What day and hour to run the weekly knowledge aggregation digest.

Aggregation per-run cost cap

Maximum to spend on AI calls per aggregation run. Deep-dive extraction stops when the cap is reached.

Deep-dive session limit

Maximum number of notable sessions to run the expensive transcript extraction on per aggregation run.

Digest sections

Choose which sections appear in the generated knowledge aggregation digest notes.

Canonical file

Which filename Yardmaster reads as the source of truth for agent instructions sync. CLAUDE.md by default; switch to AGENTS.md if that is your primary tool.

Mirror to

Pick which agent-instruction filenames to write alongside the canonical file. The canonical file itself is hidden, no self-copies.

Git guardrails enforcement level

Whether a blocked git action is hard-blocked or just warned about. The full git-guardrails controls also live in the Dev Pipeline panel.

Git guardrails protected branches

Which branch names agents may not commit, push, or edit on directly. Defaults to master and main.

Git guardrails ready-to-merge push gate

Optionally require a ready-to-merge tag before an agent can push a feature branch. Off by default.

Git guardrails

Master on/off for git guardrails: block every spawned agent from committing or pushing to a protected branch (master/main), editing while on one, or a destructive git stash. On by default.

Pause git guardrails

Temporarily let every blocked git action through, for you and every spawned agent, for a few minutes with a live countdown. Human-only.

Git guardrails excluded repos

Per-repo exclusion list: repos on this list are fully exempt from git guardrails. Human-only.

Enable Memory Palace

Persistent cross-session memory, search past sessions and auto-recall context

Enable Claude Skills integration

Show the Skills entry in the sidebar and allow skills to be injected into new sessions. Turn off to hide the integration entirely.

Enable Cron Jobs

Show the Cron Jobs entry in the sidebar for scheduling recurring tasks. Turn off to hide the feature and stop scheduled jobs from running.

Enable Recipes

Show the Recipes entry in the sidebar for saved multi-step orchestrations. Turn off to hide the feature and stop scheduled recipes from running.

Clear archive-approval pile-up

One-click sweep that archives every silent recipe session waiting on a pending session-archive approval and dismisses the corresponding approval rows.

File Converter

In-development: a standalone file converter: drag or pick a file, choose a target format, and convert it on-device (documents, images, audio/video). Free, local, nothing uploaded. Works on desktop and mobile.

Mind Map

In-development: a standalone mind-mapping canvas: build a tree of ideas with the keyboard and mouse, auto-layout, save to a library, and export to JSON or PNG.

Whiteboard

In-development: a freeform Excalidraw whiteboard: sketch and diagram on an infinite canvas, save to a library, and let an AI agent read and draw on the board for you via the control server.

Canva Integration

Connect your Claude sessions to Canva for design creation, management, and export. Sessions can generate designs, search your library, export to various formats, and manage assets, all through conversation.

SMS Integration

Send and receive text messages via Pushbullet

SMS provider

Pushbullet routes texts through Pushbullet’s cloud. Native (my phone) sends and receives over your own phone’s real SIM via a self-hosted gateway app, no third-party cloud.

Alert me when SMS goes offline

Raise an inbox alert (and one notification) when SMS is on but not working, a missing/expired Pushbullet token or a sustained disconnect. Clears itself once SMS reconnects.

AI name inference daily cost cap (USD)

Maximum spend per calendar day on Haiku calls that infer a friendly name for unknown short codes (e.g. 62438 → "eBay security code"). When the cap is hit, inference pauses until the next day. Range: $0.05 to $5.00.

Webhook Integration

Receive messages from external services like CI/CD pipelines, monitoring tools, or custom apps by giving them a URL to send data to.

Zapier Integration

Connect your Claude sessions to 8,000+ apps via Zapier. Create your MCP server at mcp.zapier.app, then paste the server URL below.

Zapier MCP Server URL

The unique URL from your Zapier MCP server. Find this at mcp.zapier.app after creating a server.

RSS Integration

Subscribe to RSS and Atom feeds in the sidebar

Add RSS Feed

Subscribe to a new RSS or Atom feed by URL

Subscribed Feeds

View and manage your RSS feed subscriptions

Article Retention

How long to keep RSS articles before automatic cleanup

Daily Digest

AI-curated morning briefing summarising your RSS feeds

Digest Delivery Time

Time of day the daily digest is delivered

Your Interests

Personalise the digest by listing topics you care about

Slack Integration

Monitor Slack channels and DMs in the sidebar

Slack Connection Status

Check Slack Socket Mode WebSocket connection

Slack API Tokens

App-Level Token and Bot Token for Slack API access

Monitored Channels

Select which Slack channels to monitor for new messages

Personal Slack Account

Send messages as yourself using OAuth instead of as the bot (requires bot setup first)

Personal Account Connection

Connect your personal Slack account to send messages as yourself instead of the bot

SMS Connection Status

Check Pushbullet WebSocket connection and device online status

Pushbullet API Token

API token for authenticating with the Pushbullet service

SMS Device

Select the Android device to send and receive SMS through

SMS Monthly Usage

Track monthly SMS sent count and free tier limits

GIF search key (Klipy)

Optional Klipy API key for GIF search in the SMS composer, GIF search works out of the box without one

Native SMS (your phone)

Self-hosted native SMS over your phone’s own SIM: phone address, username, password, and the inbound webhook URL + shared secret to paste into the SMS Gateway for Android app

SMS Favorites

Saved people you can text from the Control Space SMS tab, a name is just a label, only the number is texted; everyone you have texted already shows up automatically, add favorites here to pin the ones you use most

Add Webhook Source

Create a new webhook endpoint to receive messages

Webhook Sources

View and manage configured webhook sources

Channels

Inbound notification sources that surface in the Yardmaster inbox: SMS, Slack, RSS, Webhooks, and GitHub.

Support Chat

In-app chat for contacting support directly from Yardmaster.

Enable Daily Digest

Generate a daily briefing summarising AI session activity across all projects.

Enable Weekly Summary

Generates a summary every Monday at the scheduled hour. Disable to pause future runs; existing summaries stay visible.

Gmail Integration

Show Gmail in the sidebar for reading and sending email

Threads Per Page

Number of email threads to load at a time (10-100)

Auto-Mark as Read

Automatically mark email threads as read when you open them

Inbox Zero Celebration

Show a brief animation when you clear all emails from your inbox

Auto-Focus Reply Box

Place cursor in the reply textarea automatically when you open an email. Off (Superhuman/Gmail-style) keeps focus on the message body so triage shortcuts (E archive, S star, U unread, ! spam, X select, H snooze) work, press R to start typing a reply.

Classic Email Cleanup Layout

Use the original flat sender list for Email Cleanup instead of the AI-guided walkthrough. Turn on if you prefer the simpler view.

Block remote images in emails

Stop email senders from loading external images when you open a message. Remote images quietly report your IP address and that you opened the email back to the sender; blocking them keeps that private. Inline and attached images still show. Off by default.

Google CLI Backend

Which command-line tool Yardmaster uses for Gmail and Calendar. Defaults to gog (the backend installed and signed in on this machine today). gws is the official Google CLI, fully wired, but it needs its own separate Google sign-in before you switch to it.

Daily Digest Schedule

Choose the hour at which your daily digest is generated each day

Daily Digest Sources

Choose which data sources are included in your daily briefing

Greeting

Show or hide the greeting at the top of your daily digest

Statistics

Show or hide the daily statistics overview

Alerts

Show or hide the alerts section in your daily digest

Projects

Show or hide the per-project activity section

Communications

Show or hide the communications section (SMS, email, chat)

Calendar Events

Show or hide upcoming and past calendar events

RSS Feeds

Show or hide your RSS feed digest

Action Items

Show or hide AI-suggested action items

Suggestions

Show or hide AI-generated suggestions

Feature Ideas

Show or hide AI-detected product feature ideas

Today's Schedule

Show or hide today's calendar agenda

Daily Digest Excluded Projects

Choose which projects to exclude from your daily digest

Scheduled hour

Hour-of-day (local time, Monday only) when the weekly summary generates.

Daily spend cap

Maximum spend per day on weekly summary Haiku calls before generation pauses.

Last generated

Timestamp and cost summary of the most recent weekly summary generation.

Run now

Trigger an on-demand weekly summary generation.

Suppressed Alerts

Alerts you have permanently dismissed via "Don't show again", they will not appear in future digests until you remove them here

Email Summarizer

Auto-summarize forwarded emails

Default summary prompt

Generic prompt used when no rule matches

Default summarization model

Model used for summaries

AgentMail inbox ID

Inbox to watch for summarizable emails

Auto-archive after summary

Archive in AgentMail after a successful summary

From pattern

Match the sender address on a per-rule basis

Subject pattern

Match the email subject on a per-rule basis

List-ID pattern

Match the List-ID header, newsletters and mailing lists

Reply with summary

Reply to the original email with the generated summary

Source label

Gmail label to watch for threads to summarize

Gmail Label Summarizer

Auto-summarize Gmail threads tagged with your label

Daily cost cap

Combined daily spend limit for the email summarizer

Email Inbound

Receive emails via AgentMail, process them with Claude, and send replies

Inbox ID

AgentMail inbox identifier for receiving emails

Approved Senders

Allowlist of email addresses permitted to send messages to your inbox

Prescreen Rules

Full classifier prompt for inbound emails (default shown when empty; JSON output format always enforced)

AgentMail Inbox ID

AgentMail inbox identifier, used for receiving inbound email AND for sending via AgentMail-routed forward_email automations

Inbox Secret Token

Approved senders must include [MC:token] in the email subject. Emails without the correct token are silently dropped.

Gmail Connection Status

Check if the gog CLI is installed and authenticated

Default Inbox Query

Default Gmail search query used when loading the inbox

Weekly Summary

Monday-morning Haiku recap of the past week, stats by project, top features, week-over-week deltas, and short coaching suggestions. Lands as a briefing inside the Daily Digest inbox surface.

Email & Summaries

Umbrella section grouping the email pipeline: connect Gmail, accept forwarded mail, match summarizer rules, and roll up the daily digest.

Drive Integration

Browse, upload, and manage Google Drive files from the sidebar.

Auto-import pasted Google Doc URLs

When you paste a Google Doc or Slides URL into chat, fetch its contents instead of pasting the URL as text. Hold Shift while pasting to bypass.

Calendar Integration

Show upcoming calendar events and get reminders before meetings in the sidebar.

Sheets Integration

Browse and edit spreadsheets from the sidebar.

Google Docs Export

Right-click a project doc, scratchpad, or agent message and publish its Markdown to a Google Doc

Google Docs Template Doc ID

When set, each publish appends a new tab to this Doc instead of creating a new one

Google Docs Destination Folder ID

When set (and no template is configured), new Docs are created inside this Drive folder

Google Workspace

Umbrella section grouping Google services: Calendar, Drive, and Sheets, all on one shared sign-in.

Google Account

Sign in once to connect Gmail, Calendar, Drive, and Sheets, or disconnect your Google account

Sheets Connection Status

Check if Google Sheets is connected

Trusted Sheets

Manage which spreadsheets can be edited and restrict editable columns

Task View columns

Configure the Task View sheet tab name and column letters: section, due date, subject, URL, completion date, and the categories tab name and column

Calendar Connection Status

Check if the gog CLI is connected to Google Calendar

Calendars to show

Choose which of your Google calendars appear in the calendar view, sidebar, and digest

Drive Connection Status

Check if Google Drive is connected

Dropbox Integration

Browse, upload, and manage Dropbox files from the sidebar.

Dropbox Connection Status

Check if Dropbox is connected

OneDrive Integration

Browse, upload, and manage your Microsoft OneDrive files from the sidebar

OneDrive Connection Status

Check if OneDrive is connected

Microsoft Client ID

Your Azure App Registration application (client) ID for OneDrive

Enable Mobile Access

Reach this app from your phone or a browser on another device

Server Port

Port for the local HTTP server (1024-65535)

Bind Interface

Which network interface the mobile-access server listens on

Access Token

Authentication token for mobile-access connections

Remote Tunnel

Access from anywhere via HTTPS using Tailscale Funnel, auto-start when Yardmaster opens

Freeze-proof mobile connection

Keeps your phone connected while the app is briefly frozen, by answering your device from a separate thread. Falls back automatically if it cannot start. Takes effect after a restart.

Your Access URL

The public URL for reaching Yardmaster remotely from a phone or another device.

Local Network URL

The local-network address for reaching Yardmaster from another device on the same Wi-Fi or LAN.

Firebase Service Account Key

Service account credentials for Firebase Storage and Hosting

SSH Remote Connections

Manage SSH profiles for running sessions on remote machines

Require approval for AI session spawn

Off by default, trusted AI orchestration runs without friction. Turn ON if you want a human-in-the-loop checkpoint before any AI-initiated session spawns.

Require a source session for command-line spawns

When on, a session spawned over the command-line API is rejected unless the caller says which session requested it (the X-Yardmaster-Source-Session-Id header, Yardmaster-spawned agents have it as $YARDMASTER_SESSION_ID). Keeps every spawned session traceable to its origin. On by default, turn off to allow anonymous script or manual spawns.

Enable Agent-Driven Sessions

Allow external scripts to spawn Yardmaster sessions and send follow-up messages on their own. Off by default, leave it off unless you have a script you trust to do so.

Enable Bookmarks CLI

Allow external agents to list, create, update, delete, reorder, and launch bookmarks over the local CLI control server. On by default. Flip off to disable all bookmark CLI access at once.

Enable AI Coaching CLI

Allow external agents to read your AI coaching artifacts (interview prompts, drafts) and start coaching interviews over the local CLI control server. On by default. Flip off to disable all AI coaching CLI access at once.

Let agents bring Yardmaster to the front

When OFF (default), an agent's request to focus or pop Yardmaster to the front is ignored while you're working in another app, instead you get a dismissible inbox note naming the session that tried. When ON, agents may pull Yardmaster to the front from behind. Either way agents can still drive the UI while you're already in Yardmaster (settings tours, tutorials), and your own shortcuts, notification clicks, and phone taps are never affected.

Let finished sessions archive themselves without approval

When ON, a session that has finished its work can archive itself immediately, even if "Require approval for session pause / snooze / archive" is on, with nothing landing in your inbox. Only a session archiving itself is exempt; archiving another session over the CLI still asks. Off by default.

CLI Control Token

Authentication token for creating sessions via the local CLI server

CLI Control Documentation

Setup guide for scripting, browser bookmarks, and third-party integrations

Approval requirements

Per-action toggles for whether an outside CLI caller waits for your inbox approval or applies immediately

Always-allowed actions

Specific actions you chose to always allow from an approval card, each removable to go back to being asked

Max Turns Per Agent-Driven Session

Hard ceiling on follow-up turns for a single agent-driven session. Server-enforced.

Max Dollars Per Agent-Driven Session

Hard ceiling in USD on the cumulative cost of a single agent-driven session. Server-enforced.

Max Concurrent Agent-Driven Sessions

Global concurrency limit across all active agent-driven sessions. New spawns are refused beyond this count.

Recent CLI Activity

Audit log of actions and reads performed through the local CLI control server

Deploy Profiles

Tell the AI which account to use when deploying to Firebase, Vercel, Netlify, and other cloud services

Dry-run mode

Log what the daemon would auto-merge each tick, but do not actually spawn Claude sessions. Useful for previewing behavior before going live.

PR Merge Queue

Bind GitHub repos to Yardmaster projects so the PR Merge Queue can triage open PRs and spawn merge sessions

PR Merge Queue Merge Prompt

Edit the exact instructions sent to a single-PR merge session, with reset-to-default

PR Merge Queue Batch Prompt

Edit the exact instructions sent to a batch merge session, with reset-to-default

PR Merge Queue Trust Level

Pause-and-ask (default), fast-lane-only auto-merge, or fully autonomous merging per repo

PR Merge Queue Risky Paths

Glob patterns whose changes push a PR into the Risky lane, never auto-merged

PR Merge Queue Base Branch

Base branch (e.g. main) PRs are compared against, controls which PRs the merge queue considers

PR Merge Queue Merge Method

How Merge on GitHub lands a PR for each repo: squash (default), a merge commit, or rebase

PR Merge Queue Auto-Merge Daemon

Periodically poll repos and auto-spawn merge sessions for fast-lane PRs on fully autonomous repos

PR Merge Queue Auto-Merge Poll Interval

How often the auto-merge daemon re-checks each repo for new fast-lane PRs (1-120 minutes)

PR Merge Queue Daily Budget

Daily spend cap (USD per repo, UTC day) for daemon-spawned merge sessions. Skip when exceeded.

AI PR summaries

Generate a plain-English one-line summary of each open PR (what it does), shown as the PR's headline in the queue. Turn off to show the raw PR title only.

Connected Tools

Check and install the developer tools that Yardmaster integrates with

Auto-update Tools

Silently update Claude Code, Playwright, AgentMail, Repomix in the background; Git and GitHub CLI notify only

Tool update reminders

Periodically check your recommended tools and remind you when an update is available, one click updates it in a terminal (Claude Code, Git, GitHub CLI) or quietly in the background (Playwright, AgentMail, Repomix). Off by default

Plugins

Extend the app with community and custom plugins

Plugin Marketplace

Browse, discover, and install plugins from the marketplace

Allow unverified plugin backends

Let third-party (marketplace) plugins run a backend with full system access. Off by default.

Marketplace Admin

Admin review queue for plugin submissions

OpenClaw Gateway

Connect to an OpenClaw Gateway by WebSocket URL with a token or password to use alternative AI providers alongside Claude

Port your OpenClaw setup into Yardmaster

Bring in your existing OpenClaw setup, port your gateway connection, scheduled jobs, and recipes into Yardmaster.

Enable AutoHotKey Integration

Adds an AutoHotKey project with a built-in code editor

Word Wrap

Wrap long lines in the editor

Minimap

Show code minimap on the right side

Reload on Save

Automatically reload the running AHK script after saving

AutoHotKey Script File

Path to your main .ahk file

AHK Editor Font Size

Font size for the AutoHotKey code editor

AutoHotKey Executable Path

Path to the AutoHotkey executable used for reload

Recover hidden trailing answers

Fixes a rare case where a late-returning background tool added an "Extended thinking" line after the agent's real answer, which pushed that answer into the collapsed activity section and left only a short acknowledgement visible. Off by default (the clean folded view), turn it on to recover those answers instead.

PRs assigned to me in Inbox

Surface GitHub pull requests waiting on you, review requested from you, plus your own PRs that have changes requested, as rows in the unified Inbox. Requires the Pull Requests tab (above) to be enabled and the GitHub CLI (gh) signed in. Off by default; checks GitHub in the background every few minutes only while on.

Auto-fix master debt (vs. ask first)

When on, Yardmaster automatically spawns fix sessions for accumulated debt on master. Off (default) drops an inbox card asking you first. Requires the Master-Debt Auto-Fixer (above) to be enabled.

Auto-fix a broken cloud build workspace

When the cloud test-offload workspace fails to build (a dependency/toolchain break the cloud can’t self-heal on its own), Yardmaster raises a loud inbox card. ON (default) ALSO launches a fixer session automatically, hands-free, so the cloud self-heals without a click. Turn it OFF for a one-click card instead, a fix session costs money, so the off switch lets you approve each one.

Use the new session engine

Runs the rebuilt session engine (V2) behind your session list, switching, and history. On by default, it behaves identically to the old engine but is built on cleaner internals. Switch it off only if you notice a problem, then restart the app to apply (the engine is chosen when the app starts).

Auto-capitalize task titles

Capitalize the first word and the start of each sentence as you type a task, skips URLs, file paths, and abbreviations. Matches the behavior of the Vault notes app.

Let AI co-drive my browser tabs

Allow an agent to drive the same browser tab you are on, your real logged-in session, with turn-taking. Off by default.

Projects House Rules never reads

Projects House Rules stops reading. Turning one on deletes the corrections already collected from it.

TypeSafe API key

Key for the System One model that shadow-evaluates the waiting detector and question detection. Recorded verdicts never act on a session.

Scan interval

How often the free deterministic friction scan runs (1-168 hours).

Deep-run daily spend cap

Maximum the opt-in deep analysis run can spend per calendar day before it is refused (it spawns a paid session).

Run the deep analysis automatically every week

When on, the deep analysis runs on its own once a week (Monday morning). It spawns a paid session and respects the daily spend cap above.

House Rules lessons pass

How often the lessons pass may run once you turn it on in the House Rules view. Every two weeks skips every other week.

Labs indicator badge (example)

Example in-development feature: shows a small "🧪 Labs" badge in the app so the gating mechanism can be proven end to end.

Advanced system-wide memory reclaim

In-development: on deep free-RAM pressure, run a Windows-only system-wide reclaim (empty all working sets + purge the standby list) via a one-time admin-approved helper. May briefly reduce performance. Hidden until shipped.

Chrome extension dev (Autopilot)

When your project is a Chrome extension, gives the agent the agent-browser CLI workflow to autonomously load, drive, screenshot and reload the extension in a real Chrome, building and testing it without you clicking around. Includes a guided "Build a Chrome Extension" mission.

Playworks

In development: describe a game and have it built overnight, with a Build Report and a link in the morning. Play it in the app and point at problems, let people you share it with point at spots too, or chat with its build session to change it, then publish the changes. Off by default.

Tasks

In-development parallel task system: a keyboard outliner whose rows render as real markdown. Hidden until shipped.

Tasks v2 delegation (AI does the task)

In-development: hand a Tasks v2 task to a background AI agent that drafts it and stops for your review, surfaced as a per-task delegation chip plus a draft-review modal. Hidden until shipped.

AI Coaching

Personalized coaching sessions with structured interviews, artifacts, and progress tracking.

Recurring Special Events

In-development: add birthdays, anniversaries, and holidays that create Google Calendar events plus lead-time reminders (day-of, a week before, a month before, or any number of days before). Hidden until shipped.

Local Chat

In-development: chat with a local AI model running on your own PC via Ollama: free, private, offline. Includes a guided setup that detects Ollama, recommends a model for your machine, and downloads it.

Voice conversation (Layer 2)

In-development: talk back and forth with an AI assistant that has context about what is on screen, can grab more context, navigate the app, and (later) take actions on your behalf. Read-only in this build.

Voice barge-in (experimental)

In-development: interrupt the spoken AI response by talking over it. Reveals the barge-in controls + mic calibration in Voice settings. Not yet fully wired, so hidden by default.

Voice "about to act" highlight

In-development: when voice asks you to confirm an action (like sending a message or ending a thread), highlight the on-screen thread it will act on so you can see what you are confirming.

Built-in free voice typing

In development: speak to Yardmaster with no API key. A small speech model runs on your own computer, free. When it is on, the voice settings gain a "Built-in (free)" speech option, and speaking works out of the box even before you add any key.

Flowcharts

In-development: a standalone visual flowchart editor: shapes and connectors, swimlanes, auto-layout, a saved library, and AI sessions that can read and draw on the board.

Feature Discovery Nudges

A daily inbox nudge that pitches an Yardmaster feature you have not used yet, with a one-click "Start using it".

Trim unused snooze times

In development: drop the snooze durations you never pick from the menu (after about a month), keeping the rest in the standard order. Off shows the full list.

Clipboard history

In-development: a clipboard-history button in the chat composer that keeps your recent copies (text + images) so you can re-insert them. Captured in memory and cleared when Yardmaster restarts, unless you opt into keeping it across restarts (saved encrypted on this PC).

Stuck-task helper

In-development: notices when you keep snoozing the same session, email, PR, or message and, after the 3rd time, offers a dismissible inbox nudge to start an AI session that interviews you about the resistance and helps you take a tiny first step.

Daily Journal

A daily inbox nudge to run a short journaling check-in (and a weekly review) with your AI coach. Entries stay private to coaching and never leak into your other sessions. Requires AI Coaching.

Native slash commands

In-development: type slash commands like /model, /clear, /cost, /context and /status in the chat composer and have them run for real in Claude Code (switching the model live, clearing context, or showing real cost/usage) instead of being sent to the model as text.

Unified sidebar engine (V2)

In-development internal rebuild: the project sidebar and Needs-You render through the same engine the global Inbox already uses, instead of a separate session-row + navigation path. No visible change when on; it exists to prove the rebuild matches today exactly. Hidden until shipped.

Master-Debt Auto-Fixer

In-development: watches the master branch for accumulated debt (failing checks, flagged findings) and either drops an inbox card asking you to fix it, or, when you turn auto-fix on, spawns fix sessions for you. Hidden until shipped.

Browser Logins

In-development: log into a website once and every agent you run can use that login automatically, many at the same time, until the site signs you out. Local-only; no passwords are stored. Hidden until shipped.

Plugin UI access

In-development: let installed plugins add their own buttons to the header toolbar and session menus, and navigate the app to a session, project, or view. Gated until the capability is proven: off by default, and enforced in Main so the whole capability stays inert until you turn it on.

Dev Pipeline

In-development: a projects-bar panel to oversee the dev-pipeline workflow: see every active pipeline run and which gate it is waiting at, turn gate auto-approval on/off, and control the auto-lander plus a history of every branch it has landed. Hidden until shipped.

Weekly Session Analysis

In-development: a developer-tools panel that mines your past agent sessions for friction patterns (command timeouts, hook blocks, tool errors, redundant directory changes, and spend concentration) via a free periodic scan. An opt-in WEEKLY deep run spawns a paid AI session to produce a full analysis report with ranked fixes. Hidden until shipped.

House Rules

In-development: learns the standing rules you keep repeating to your agents. It collects your corrections from past sessions on this computer, groups the ones that recur, and once per cycle asks one AI session (on your own subscription) to propose short rules for you to approve. Hidden until shipped.

Workflows

In-development: build n8n/Zapier-style automations on a visual canvas: a trigger plus a graph of steps (deterministic actions and AI steps) that Yardmaster runs for you.

Share Comments

In-development: commenting and feedback on shared artifacts. Viewers can leave inline or general comments, which cross-post to Team Chat and the inbox.

Plugin AI-native CLI

In-development: lets the Yardmaster AI and CLI invoke an installed plugin's own actions through its registered CLI endpoints (ctx.cli.handle). Off by default until the host dispatch path is proven. No UI surface; it gates a control-server route.

New Settings layout

In-development: reorganizes Settings into clearer, better-named "rooms" without changing any individual setting. Off by default; ships hidden and is filled in over follow-on updates until it is ready to try.

Keyboard Demo tour

In-development: a keystroke-driven copy of the Interactive Demo. Each step shows a key to press (the app’s real shortcuts), so the walkthrough doubles as hands-on keyboard practice on safe sample data. Reveals a "Keyboard Demo" card in Settings → Setup Wizards.

Voice attachment indicator

In-development: a small chip showing whether voice targets Yardmaster (app-level) or a specific session, plus a per-utterance receipt of where each voice command actually went.

Spoken Narration

In-development: a spoken-audio recap of each completed reply, the audio sibling of Plain Speak. When on, the agent writes a short spoken script that plays back per message (optionally auto-playing new replies). Hidden until shipped.

Inbuilt terminal

In-development: a real shell terminal inside Yardmaster, scoped to a project, powered by xterm.js + node-pty. Off by default until it ships.

Terminal command blocks

Show visual command boundaries (check / cross per exit code) with fold, copy, and re-run per block. Requires the inbuilt terminal to be on.

AutoHotkey Manager

In-development: a no-code manager for AutoHotkey hotstrings, hotkeys and snippets. Yardmaster generates and maintains the .ahk file for you.

AI-drivable browser

In-development: lets any Yardmaster session drive the embedded Browser with AI (navigate, read the page, click, type, screenshot) over the CLI control server (Cursor-browser style). Hidden until shipped.

Repo Foundations

In-development: a "Set up this repo" card on the project dashboard that runs the Repo Foundations skill to install a tailored documentation system + Claude-rules baseline (after a preview you approve). Spawns a real paid session. Hidden until shipped.

Dependency Security Scan

In-development: warns you when a project depends on a vulnerable or malicious package. Hidden until shipped.

Portable Backup

In-development: export an encrypted backup file you can move to another computer, and import it back. Hidden until shipped.

Per-project worktree location

In-development: a "Worktree location" control in Edit Project → More options that sets where this project's git worktrees are created, saved as a private per-machine override that never touches your git history. Hidden until shipped.

Picovoice wake-word engine (retired)

Retired: the paid Picovoice wake-word engine. Its proprietary models are no longer bundled and it is hidden from the wake-word engine picker; the free built-in engine recognizes the same wake words. Turn on only to re-reveal it for testing.

GPT & Grok via model proxy

In-development: run GPT or xAI Grok inside the Claude Code harness by pointing it at a local model-translating proxy you run yourself. Off by default until the bundled proxy ships.

Plan limits

In-development: enables per-plan resource limits (today, only the project count cap). The current version deliberately enforces no limits: every install is unlimited until this ships alongside billing/plans.

The Board (dispatch UI)

Switches the main screen to The Board: a dark dispatch panel where the chrome carries no color and the only lit thing on screen is a session telling you what it is doing. Classic stays available; switch back any time.

Blueprint

In development: answer a guided interview and Blueprint researches, extracts your design system, and writes a complete brief you can paste into Claude Design.

Waiting-detector shadow evaluation

In development: ask a System One model (TypeSafe Jev) what it would have decided, record the answer, and act on none of it. It is asked on a turn the waiting detector is unsure about, and on every finished turn whether the agent is asking you something. Purely observational, so sessions are routed exactly as they are today. Needs a TypeSafe API key and costs about 3 cents per thousand decisions.

Slack sweep

In development, owner only: connect Slack with your own browser sign-in so a scheduled sweep can catch you up on the channels you choose. Only the owner of this workspace can connect or disconnect it.

Slack Posts

In development, owner only: write Slack posts with AI and send them as you on a schedule, after a test post and your approval. Only the owner of this workspace can use it.

Re-imagine a Share

In development: ask AI to rethink how a Kit Share looks (house, layout and block choice) without changing a single fact, within a daily limit you set. Off by default.

Share Brain

In development: Yardmaster notices when a file an agent just wrote is worth sharing and offers to publish it. Off by default.

Junction

Send one request to several OpenRouter models and pick the best answer blind.

Roundhouse

A team of AI agents with roles that works through a shared ticket board.

Test Track

Put one task to several models and compare how each one does.

Prompt A/B

In development, owner only: give each new session one of two wordings of three system prompt sections, measure both, and keep the better one. Only the owner account of this app can start or stop it.

Live model catalog

In development: fetch the owner-published list of newly released models so they appear in the model picker without an app update. Built-in models are never changed.

Send when done

In development: queue follow-up messages for a session and have each one sent automatically when the session finishes its turn.

Reader questions to your assistant

In development: let readers of a public Share ask a question that this app answers from the page, on your own AI route and within your daily limits.

Automatic transcripts

In-development: Yardmaster turns a video's speech into a transcript on your Mac, for free.

Setup Backup to Gmail

Mails an encrypted bundle of quick replies, automations, recipes, bookmarks, keybindings, and skills to your Gmail address weekly

Encryption passphrase

Optional passphrase used to encrypt the setup backup bundle

Restore from backup

Restore quick replies, automations, recipes, bookmarks, keybindings, and skills from a setup backup file

Backup Mirror

Encrypted off-machine backup of the full database, config, and attachments so session history survives a hard drive failure

Mirror folder

Destination folder where encrypted mirror archives are written, typically a synced cloud folder

Mirror passphrase

Passphrase used to encrypt and decrypt the mirror archive, required to restore on a new machine

Mirror retention, keep last

How many mirror archives to keep last before older ones are pruned

Mirror automatic sync

Keep two devices in step automatically, on a schedule (sync every hour, 3, 6, or 12 hours) or when you open and close Yardmaster

Mirror actions

Run a mirror immediately, refresh the archive list, or restore an existing archive (merge or replace)

Automatic Backups

Automatic local backups on startup and periodically

Export Data

Export all app data to a backup ZIP file

Import Data

Restore app data from a previously exported backup

Erase a sender's emails

Permanently delete every stored inbound email from one sender, for a data-subject privacy request

Portable Backup

Export an encrypted, self-contained .yardmasterbackup file (or import one from another computer), a single file with a recovery code, no cloud account or matching passphrase required

Post-restore credential check

After restoring from a backup, check which saved credentials, API keys, CLI logins, SSH remotes, and toolchains need re-entering, then fix any that broke because they were encrypted with the previous machine keys

Cloud backup

Store your encrypted portable backup in the cloud, tied to your account and end-to-end encrypted; restore it on another machine with your recovery code

Log Level

Controls how much detail is written to the log file. Higher levels include all lower levels. Use "Info" for normal operation or "Verbose" when troubleshooting.

Performance & interaction tracing

When on, Yardmaster records freeze/long-task and click-latency timing to local troubleshooting logs (renderer diagnostics + interaction.log). Turn it off to stop that logging and the overhead it adds to the main process, your crash reports, error logs, and the freeze recorder stay on. Nothing is sent anywhere.

Send automatic problem reports to the Yardmaster team

When on, the one-click "Send report" buttons (a file link that points nowhere, a snooze phrase that was misread) send a short report to the Yardmaster team. Off stops those. Feedback you send yourself from the bug icon always goes.

Send usage numbers

While you are signed in, Yardmaster sends a numbers-only usage report (sessions, AI cost, tokens, active time and similar totals) to the Yardmaster team about every 20 minutes. It is tied to your account and never contains project names, session titles or anything you typed. Off stops it.

Send error reports

When something goes wrong, Yardmaster sends a text error report (the error, where it happened and recent app activity, with secrets and email addresses removed) to the Yardmaster team through Sentry. Crash dump files are never uploaded; they stay on this computer. Off stops all error reports at once. Turning it back on takes effect the next time Yardmaster starts.

Reclaim space from old archived sessions

When on, Yardmaster clears the hidden sub-agent (Task helper) output from sessions you archived over 14 days ago. Your conversations are untouched and a full backup is taken first; you just lose the raw text of those collapsed "subagent result" cards in old archived sessions.

Track which UI controls you use

When on, Yardmaster records which buttons, menus, and sidebar integrations you actually use, so you can see what is used vs never used in the "UI Usage" panel below and decide what to remove or hide. The report stays on this device and is never sent anywhere.

Automatically tidy away controls you never use

When on, Yardmaster moves sidebar integrations and toolbar icons you have not used in 30+ days into a quiet spot (an "Unused" group / the "…" overflow) to reduce clutter, and tells you in the inbox. Nothing is deleted, restore anything in the list below. Off by default.

Pre-compute codebase stats in the background

When on, Yardmaster quietly analyzes and caches each project’s codebase stats (size, languages, git activity, code health) in the background, so the "Codebase Stats" view opens instantly instead of computing on the spot. It is heavily throttled, one project at a time, only what changed, and it backs off when your computer is busy, so it never slows things down even with many projects. (Dependency security scans still run only when you open a project and refresh.)

Warn me about low disk space

When on, you get an inbox warning if free disk space drops below the threshold below and automatic cleanup cannot free enough. Turning it off silences the warning only. The automatic cleanup that protects your disk keeps running.

Warn when free space drops below

The amount of free space to keep on your working drive. Below this, finished worktrees are cleaned up automatically, and you are warned (when the warning above is on) if cleanup cannot free enough.

Usage numbers and error reports

Turn off the usage numbers and the error reports sent to the team, each with its own switch under Reports to the team. Usage numbers are sent only while you are signed in and are tied to your account. Crash dump files are never uploaded

Log Files

View log file location, sizes, and contents

Export Logs

Download a ZIP of all log files and crash dumps for support

Recent Errors

View recent errors and warnings from the application log

Crash Dumps

Native crash dump files from GPU or renderer process crashes

Settings Recovery

Shows whether your settings file loaded cleanly or was restored from a backup at startup

Startup Trace

Self-contained startup performance log for bug reports

Interaction Trace

Timeline of slow clicks, keypresses, and rendering jank for bug reports

Main-Process Heartbeat

Event-loop lag + memory + last IPC op tape, forensic trail for silent crashes

Spawn Audit

Live audited child processes + dead instance ledger consulted by the orphan reaper

Session Turns Cache

Pre-built turn-grouped chat layout cache, coverage status and invalidate-all action

Waiting Detector activity

Audit log of every Gate-6 waiting-detector firing. Tag verdicts (👍 Good / 👎 Bad with optional reason) and export the bundle as JSON for AI-assisted pattern tuning.

Rescan project icons

Clear and re-detect all project sidebar icons

Refresh GitHub auth

Clear the cached GitHub token after gh auth switch

Resources

Live CPU and memory for Yardmaster processes and pausable services

Chat open speed (old vs new engine)

Times how long a session takes to open for the old chat engine vs the new Chat v2 engine, so you can compare their speed.

Prefer cloud for tests on this machine

Run agent test commands (tests, typecheck, lint, E2E) on the shared cloud VM instead of locally when your work is committed. Off by default; affects only this machine.

Cloud-only builds on this machine

Run the verification build on the shared cloud VM instead of this computer; build-affecting uncommitted changes ask you to commit first instead of building locally. The final release build always stays local. Off by default; affects only this machine.

Hidden items

Items Auto-Tidy has hidden from the sidebar, restore any of them here

UI Usage

Which controls and integrations you use vs. never use, so unused things can be removed or hidden. Stays on this device.

Flow

What you tend to do right after a given control, reconstructed from local usage sequences on this device. Stays on this device.

Apply updates automatically when away

When an update is ready, install and restart automatically only if no sessions are running and your computer has been idle. Off by default.

Version History

View release notes and changelog for all app versions

Share defaults moved to the Shares panel

Theme, comments, expiry, saved state, live updates and alerts for new shares

Terminal snippets Preview

Manage saved commands for the terminal snippet palette

Still in development and off until you turn it on: turn on Inbuilt terminal (Settings > Lab) first.