Collaboration

Real-time team collaboration on agent tasks — one workspace, one thread, zero version confusion

What is Collaboration?

Opulent Collaboration brings your entire team into the same agent workspace. Instead of one person running an AI task and then sharing results via email or Slack, your team works together in a shared thread — everyone sees the agent working in real time, anyone can add context or redirect, and the output belongs to the whole team.

This eliminates the coordination overhead that makes AI tools a bottleneck rather than a multiplier: no copy-paste of results, no "which version is current", no parallel agents producing conflicting outputs.

The Problem It Solves

When AI is a solo tool:

1. One person opens Opulent
2. Runs a task
3. Exports or screenshots the result
4. Pastes it into Slack or email
5. Teammates have questions → repeat the task with updated context
6. Two people run similar tasks → conflicting outputs → confusion

With Opulent Collaboration:

• The thread is shared — everyone sees the agent live
• Anyone can add a message to redirect or refine
• The output is in one place, with full history
• No version confusion, no copy-paste, no context loss

How It Works

Invite to a Thread

Open any task thread and click Share. Invite teammates by email or workspace member. They can join immediately.

Shared Context

All participants see:

• Every message in the thread
• The agent's tool calls and outputs as they happen
• All documents, slides, sheets, and files produced
• The full conversation history

Real-Time Participation

While the agent is working, any team member can:

• Send a follow-up message to refine the direction
• Upload additional context files
• Ask the agent to focus on a specific sub-task
• Approve or reject a direction before the agent continues

Output Ownership

Documents, slides, and data produced in a shared thread are available to all participants. No one needs to "send" the file — it's in the shared workspace.

Collaboration in Practice

Strategy + Execution Loop

Thread: Q4 Go-to-Market Planning

Strategy lead: "Research the enterprise AI tooling market — segments, key buyers, top 5 competitors"
Agent: [Runs parallel research → produces market overview]

Marketing: "Good. Now turn this into a 12-slide strategy deck for the leadership meeting"
Agent: [Produces slides]

CEO: "Adjust slide 4 — our differentiation point needs to be sharper"
Agent: [Updates the specific slide]

All three review the final deck together in the thread.

Document Review Workflow

Thread: Product Launch Blog Post

Writer: "Draft a 1,200-word blog post on our new AI agent platform"
Agent: [Produces draft in Document panel]

Editor: "The intro needs a stronger hook. Rewrite the first two paragraphs"
Agent: [Updates document]

Marketing: "Add a CTA at the end linking to our demo page"
Agent: [Appends CTA section]

Final document is shared — everyone approved it in the thread.

Data Analysis Collaboration

Thread: Monthly Revenue Review

Finance: "Pull our Stripe MRR for Q4, broken down by plan tier"
Agent: [Queries Stripe → populates Sheets panel]

CFO: "Flag any accounts that churned after the price increase in November"
Agent: [Identifies and highlights at-risk accounts]

Sales: "For each flagged account, find the AE owner in Salesforce and draft a win-back note"
Agent: [Cross-references Salesforce → drafts emails]

Access Controls

Thread Visibility

• Threads are private by default — only the creator can see them
• Sharing is explicit — you choose exactly who to invite
• Invitees can see everything in the thread, but cannot add other participants without owner permission

Workspace Collaboration

Workspace-level sharing gives teammates access to:

• The workspace master instructions and knowledge base
• All threads they've been explicitly invited to

They do not automatically see threads they haven't been invited to — even in a shared workspace.

Tips

Share threads before the task is complete. The most value comes from collaborating during execution, not just reviewing results.

Use mentions to direct the agent. In a shared thread, mentioning a specific colleague's context helps the agent tailor its output: "The output needs to be useful for @sarah's team, who focus on enterprise accounts".

Keep one thread per topic. Avoid splitting related work across multiple threads — the agent loses context about what was already done.

Let the agent serve as meeting notes. After a team discussion about requirements, have the agent summarize decisions and action items in a shared thread. Everyone has the same record.

Common Questions

Can participants invite additional teammates? Participants can suggest additions, but only the thread owner can approve new members to ensure sensitive tasks remain controlled.

What happens if two people send messages simultaneously? Messages are queued. The agent processes them in order.

Can I make a thread public (shareable link)? Yes. Threads can be shared via a public link for read-only viewing — useful for sharing agent outputs with external stakeholders without granting edit access.

Can I @mention the agent directly? In shared threads, any participant can send messages to the agent. There is no special mention syntax needed — all messages in the thread are seen by the agent.

Is collaboration available on all plans? Yes. Thread sharing is available on all Opulent plans.