Your agents work.
The conversation stays here.
Unl is your why agent. It keeps what you decided and why, and gives the relevant part to every AI and agent you use before it acts.
When your agents need a new decision, the conversation can happen in Slack.
Answer in words.
Keep the reason.
An agent reaches a choice it cannot make on its own. Unl brings the question into a Slack thread. You explain what should happen and why. A no without a why is a yes, so the reason goes with the answer. Your answer reaches the agent that asked.
The thread stays open for the back and forth. What the agent reports comes back to the conversation, and your next reply stays with it.
Connect Unl to Slack- needs-your-input
- build
- releases
- Unl
- 1 You answer in the thread
- Unl
Your coding agent needs a decision. Keep the existing library or replace it? Nothing you have decided yet covers it.
Needs your input - You
Keep it. The team already knows how to maintain it.
- Unl
Kept, with your reason. Your coding agent gets it on its next turn.
Answered - 2 Your agent continues, with your reason
- In your coding agentUnl One decision bears on this. Keep the existing library, because the team already knows how to maintain it.Agent Extending the existing library.
- 3 The result comes back to the thread
- Unl
Your coding agent reports back. The change extends the existing library, as you decided.
Result
Only the new judgement needs you.
- The knock
The question arrives.
Unl brings you the case and the decision it touches, in the place you already work.
- The thread
You decide it.
Ask a question, explain the tradeoff, or give your answer. The conversation carries the why.
- The next turn
Your agent continues.
Unl keeps the answer with its reason. The agent that asked gets your words on its next turn.
People and agents.
One decision conversation.
The direction is a thread where people, Unl and other agents work through the same question. Agents bring evidence, proposals and results. The person responsible for that area decides in their own words.
Direct participation by other agents is still being built. The illustration above shows the decision flow, rather than a captured multi-agent thread.
See the team modelYour why goes back into the work.
Unl is your why agent. It keeps what you decided and why, and gives the relevant part to the AI and agents you use before they act.