Lithora AI

Your intelligent work assistant powered by advanced AI.

AI Workspace

IDLE

issue #418github · open

users lose draft state on refresh; editor is slow on big docs; we should add autosave.

Waiting for work…

The AI Workspace reasons over your linked context, proposes a plan, waits for your confirmation, then executes — it can take an issue all the way to a merged PR.

Overview

Lithora AI is built into the platform to help you work smarter. It can analyze your work, suggest priorities, break down tasks, and help you stay on track.

AI Workspace

Access the full AI assistant at Dashboard → AI where you can have conversations, link context, and execute actions.

Plan requirement

AI features require an active Pro or Scale plan — both include a 7-day trial with no credit card required. An account with no active trial or subscription is locked and sees the upgrade path instead. See Billing & plans.

How AI Works in Lithora

Lithora uses two AI systems optimized for different use cases:

AI Workspace (/dashboard/ai)

Full AI agent with tools and action execution. Link tasks, projects, and files to give AI context, then ask it to analyze, summarize, or take actions.

Create and update tasks
Analyze linked items
Suggest next steps
Break down complex work

AI Features Everywhere Else

Quick AI assistance integrated throughout the platform for specific tasks.

Whiteboard AI suggestions
Project summaries & bottleneck detection
Agent actions inside automations
AI-drafted workspace pages & templates
Meeting summaries & AI standups

AI allowances

AI messages are a daily team pool — everyone on the team draws from the same allowance. AI PR reviews are granted per seateach month and pooled across the org, so adding a seat raises the whole team's ceiling.

Plan	AI messages (team pool, daily)	AI PR reviews (per seat, monthly)
Pro — $10 / seat / month	500 messages/day	5 per seat
Scale — $29 / seat / month	3,000 messages/day	15 per seat

Message limits reset daily at midnight UTC and PR-review allowances reset monthly. Neither rolls over. Custom Enterprise plans are built on Scale's entitlements, so they start from these numbers.

Whose plan applies

Paid capabilities resolve from the team owner'splan, not each member's. If your team is on Scale, every member gets Scale's AI allowances even though their individual account carries no subscription of its own.

Context Linking

In the AI Workspace, you can link items to give AI context about your work:

Tasks: Link tasks to discuss status, blockers, or get help breaking them down.
Projects: Link projects to analyze progress or identify risks.
Files: Link files for AI to reference when answering questions.
Whiteboards: Link whiteboards to discuss ideas and brainstorms.

Privacy

AI only sees items you explicitly link. Your other data remains private and is never used for AI training.

The approval gate

The agent does not write to your workspace on its own initiative. Anything that would change data goes through a confirmation gate:

The agent proposes an action and stages it as pending, describing exactly what it intends to do.
You approve it. Clicking the confirm button is the confirmation — nothing executes before that.
Lithora re-validates your permissions, executes, and records the result.
Every step — proposal, approval, execution, and any failure — is written to an audit log.

A staged action can also be rejected

Autonomous runs triggered by a webhook (a PR opening, a CI failure) follow the same path: they stage a plan and wait. If you never approve it, nothing happens. See Agents & presets for the agents that use this.

What AI Can Do

Analyze: Get insights on your linked tasks, projects, and workload.
Summarize: Quickly understand complex projects or documents.
Create Tasks: Ask AI to create tasks based on your requirements.
Update Work: Change task status, priorities, or details via conversation.
Suggest: Get recommendations for next steps or improvements.