TARS

build c96df15
TARS
Run /plan to classify and schedule these

Sign in to see your calendar

Work Sessions

⏱️ Active Session

Started: Elapsed:

Today's Sessions

No sessions logged today

Total Today: hours ( min)
Sessions:

Meeting Notes

Connect Atlassian in Settings to see your meeting notes

Loading meeting notes...

No meeting notes found

Insights
Loading insights...

Select an insight from the sidebar or create a new one.

No meeting profiles yet. Run /recap to auto-create one, or /meeting-profile [name] to build one manually.

No user model yet. Run /usermodel in chat to build one.

Unsaved changes

Connect your Slack workspace to use Slack Alerts.

Alerts

No matched alerts yet.

Create an alert to start monitoring, or select messages from the feed.

Alert Definitions

No alert definitions yet. Create one above or select messages from the feed.

Feed

Messages from monitored channels will appear here as they arrive.

Create Alert

Knowledge Graph

These channels will be skipped when reading Slack during Build. Use this to exclude noisy/automated channels.

Connect Slack to configure excluded channels.
Loading channels...
Loading graph...
Entities
Relationships
Predicate Types
Warehouse Chunks

Graph

● Top entity ● Entity

Warehouse

All
Avg Quality
Embedded
Entity-Linked
Summaries
Clear

Relationships by Type

Top Entities

Last updated:

Press Build to populate it from your Slack conversations, Jira tickets, and calendar events. This powers /people, /prep attendee context, and semantic search.

Worker LLM Log

Purge Knowledge Graph

This will permanently delete all entities and relationships. This cannot be undone.

Purge Warehouse

This will permanently delete all semantic chunks. This cannot be undone.

Scenario Moves
▸
Scenario is now the accepted plan. Draft scenarios will build on top of this.
Workforce & Skills
Import from Jira or add people manually.
Chats

Setup tips
💡 Open the KG tab and press Build to populate your knowledge graph from Slack, Jira, and calendar.
💡 Run /usermodel to build a profile of your priorities and working style.
Priority Push
🔧
Queue: pending
1 3 5 2 4 A
Switch to Opus? This is the expensive tier.
Transcribing Loading speech recognition Generating speech

Time
Location
Organizer
Meet
Attendees
Description

Rename Chat

New Chat

Settings

AI Configuration

Gear Matrix (Auto mode command defaults)

Todo Source

Choose where to fetch your tasks from.

Jira Todos

Connect your Atlassian account above to configure Jira todos.

Atlassian (Jira / Confluence)

Connect your Atlassian account to access Jira and Confluence.

Connected to

Meeting Categories

Customize categories for organizing meeting notes via the /recap command.

Slack

Connect your Slack workspace to search messages and send DMs as yourself.

Connected to

Configure which Slack scopes to search.

Slack Lists Configuration

Enter the Slack List ID to use as your todo source.

Connect to Slack above to use Slack Lists as your todo source.

Miro

Connect Miro to let the AI read your boards for research and insights.

Connected to Miro

Bitbucket

Connect Bitbucket to let the AI search repos, PRs, branches, and pipelines.

Connected to Bitbucket

GitLab

Connect GitLab to let the AI search projects, MRs, branches, and pipelines.

Connected to GitLab

MCP Servers

Grafana Loki

Query application logs via Loki. The AI can search logs, check errors, and investigate incidents.

Loki configured

Email

Team / Org Structure

Add your manager and team members so the AI can answer questions about your team, reporting lines, and org structure.

Loaded from Drive:

Tool Permissions

Tools permanently approved to execute without confirmation. Click × to revoke.

Proactive Interventions

Automatically surface priorities, stale follow-ups, and actionable items at session start.

Slack Alerts

Connect Slack above to use alerts.

Monitor channels and get alerted when messages match your alert definitions.

Select up to 20 channels. Currently selected.

Alert sensitivity is now configured per-alert in the Alerts tab.

Advanced

Edit the assistant's behavior, prompts, patterns, and tools. Changes persist to Drive.

View and restore previous versions of your data files from Google Drive.

Add Task

View in Jira ↗ Slack List

Labels

No labels

Description

Comments

Loading comments...
No comments yet.

Edit Work Session

Add Manual Session

Skill Editor

Unsaved changes

Data Backups

Data Files

Confirm Action

⚠

LLMs are bad at math — this runs code in a sandbox to get accurate results.

🔗

This action requires a service that isn't connected yet. Would you like to connect now?

Help

Getting Started

  1. Sign in with Google
  2. Open Settings (⚙) and configure:
    • OpenRouter API Key — Get from openrouter.ai/keys
    • Connect Atlassian — Jira tasks, Confluence notes, project tracking
    • Connect Slack — Search, DMs, channels, thread reading, alerts
    • Connect Bitbucket / GitLab (optional) — PRs, pipelines, code search
    • Tavily API Key (optional) — Web search from tavily.com
    • Calendar Embed URL — From Google Calendar settings
  3. Type /setup to auto-configure Slack todos + Confluence meeting notes
  4. Click Save to Drive

Slash Commands

CommandDescription
/sodStart of Day — calendar, tasks, messages, top 3 priorities
/standup [me|team]Standup summary for yourself or your team
/eodEnd-of-day summary and tomorrow preview
/weekWeekly summary
/plan [today|tomorrow]Classify tasks for day plan (deep/short/deadline)
/pulse [week|month]Work pattern analysis against research-backed targets
/work [task]Start tracking work on a task
/endEnd current work session and log blockers
/prep [meeting]Prepare for an upcoming meeting (pulls KG, Slack DMs, labels)
/recap [cat] [subcat] [name]Summarize meeting → interactive review → Confluence + Jira
/analyse [topic]Analyse patterns from meeting recaps and data
/find [query]Search across all sources
/context [project]Pull all context for a project
/research [query]Deep multi-step research with extended thinking
/research-plan [query]Plan a research investigation before executing
/discover [interests]Discover Slack channels to join based on your role
/todo [task]Quick-add a todo (syncs to Jira if configured)
/save [title]Save an insight from the current conversation
/alerts [status|learn|clear]Manage Slack alert patterns
/newStart a new chat
/clearClear current chat history
/compactCompress conversation to save context
/setupAuto-configure Slack todos + Confluence meeting notes
/skillsReview custom skill modifications and resolve conflicts
/diagnoseRun system diagnostics (check all connections)
/helpShow this help screen

Daily Workflow

A typical day with TARS:

/sodStart your day — mission briefing with top 3 priorities
/standupShare your standup — yesterday's completions, today's plan, blockers
/planClassify today's tasks — deep work, short tasks, deadlines
/work [task]Start a timer when you begin a task
/endStop the timer, log notes and blockers
/eodEnd your day — summary with tomorrow's prep
/weekWeekly rollup — hours logged, completions, carry-overs
/pulseAnalyse your work patterns — focus time, meeting load, context switches

Meeting Lifecycle

TARS separates capture (what happened) from analysis (what it means). Raw data stays useful; conclusions can evolve.

/prep [meeting]Before: briefing with your items, Slack DM context, KG relationships
/recap [cat] [sub] [name]After: transcript → interactive review → Confluence + Jira tickets
/analyse [topic]Later: cross-reference recaps, test insights, surface patterns

Meeting notes are organized in Confluence: Parent Page / Category / Subcategory / Recap

Label your todos with meeting names so /prep automatically pulls the right tickets.

Research & Context

/find [query]Search across Jira, Confluence, Slack, and Gmail
/context [project]Full context pull — issues, docs, mentions, health, KG relationships
/research [query]Deep multi-source investigation with knowledge graph + pattern detection
/research-plan [query]Plan the investigation before executing
/discover [interests]Find Slack channels relevant to your role or interests

Voice & Audio

Enable Show voice controls in Settings to use speech-to-text and text-to-speech.

STT (Whisper)Click the mic to speak. Runs locally via WASM or WebGPU.
TTS (Kokoro / SpeechT5)TARS reads responses aloud. Fully offline, no API keys needed.
Voice modeClick mic for one question. Hold for continuous hands-free.
Wake wordsSay "hey tars" to activate (enable in Settings).
Pause / ResumePause TTS mid-sentence and resume where you left off.

Sandbox (Compute)

The AI can run JavaScript in a secure sandbox for calculations and data analysis. Available libraries:

LibraryKey Functions
ss (simple-statistics)mean, median, standardDeviation, percentile, linearRegression, sampleCorrelation, quantile, interquartileRange, min, max, sum
d3 (d3-array)group, rollup, index, bin, extent, sum, mean, median, sort, ascending, descending, cross, zip
storeset(key, value), get(key), delete(key), keys() — persists across calls within a conversation (1MB limit, resets on page refresh)

Integrations

ServiceCapabilities
GoogleCalendar, Gmail search, Meet transcripts, Drive config storage
AtlassianJira (tasks, projects, epics, time logging, labels), Confluence (pages, meeting notes)
SlackSearch messages, read threads, send DMs, post to channels, alerts monitoring
BitbucketRepos, PRs, branches, commits, pipelines, code search
GitLabProjects, MRs, branches, commits, pipelines, code search
LokiQuery application logs (LogQL), discover labels and services
MiroSearch boards, extract content
Web SearchExternal search via Tavily (docs, news, how-to guides)
TimeTimezone lookups, public holidays for 100+ countries

Keyboard Shortcuts

EnterSend message
Shift+EnterNew line
UpRecall previous message
EscapeStop generating
Alt+←Navigate back
Alt+→Navigate forward

Security

  • Zero local storage — all config stored in your Google Drive
  • OAuth everywhere — Google, Atlassian, Slack, Bitbucket, GitLab, Miro all use OAuth 2.0
  • Memory-only tokens — credentials cleared on page close
  • User-owned data — your config lives in your own Drive
  • Sandboxed compute — code runs in an isolated Web Worker with allowlisted APIs only (no network, no storage, no filesystem)

Configuration

SettingDescription
OpenRouter API KeyAPI key for LLM access
Tavily API KeyAPI key for web search (optional)
ModelWhich model to use (Claude, GPT-4, etc.)
Calendar Embed URLYour Google Calendar embed URL
Todo Project KeyJira project for personal tasks
Confluence Space KeySpace for meeting notes
Parent Page IDRoot page for meeting notes hierarchy
Watched ProjectsComma-separated Jira project keys to track
Watched EpicsComma-separated epic keys to track
Watched BoardsBoard ID:ProjectKey:Name for sprint-level tracking
Custom ViewsPipe-separated Name|JQL pairs for Projects tab
Meeting CategoriesCustom lists for Projects, Teams, and 1:1s
Knowledge GraphEnable to build entity relationship graph from your tools
ScenariosEnable for what-if resource planning with Gantt charts
Slack AlertsEnable for channel monitoring with pattern-based classification

Tips

  • Press Enter to send a message, Shift+Enter for a new line
  • Use /compact when conversations get long to save context
  • Type /setup after connecting services to auto-configure
  • Label todos with meeting names so /prep pulls the right tickets
  • Insights saved via /save are treated as testable hypotheses — validated against current data
  • The Knowledge Graph learns relationships as you use TARS — no manual setup needed
  • The AI prefers the sandbox for any computation with more than a few data points
Update available (current: c96df15)