Aurora at work
AI Agents for Product Teams
Practical product playbooks for a team of individually importable Finta AI coworkers.

Build a product team around a question the evidence can answer. Move from customer feedback to a scoped feature brief, prepare a prototype learning plan or assemble a release decision. Each workflow keeps the user task, observable acceptance and the accountable owner at the center.
Choose your first job
Use the feature guide when the problem and scope need reconciliation. Use prototype planning when one interaction or comprehension question could change the decision. Use release review when an actual revision and its verification evidence need a clear disposition.
Build your starter team
| Agent template | Its job on the team | Bring to its setup | Reviewable return | Team |
|---|---|---|---|---|
Customer Voice ResearcherTeam agent | Build a traceable customer-problem evidence pack. | Authorized feedback, source context and the decision at stake. | Attributed observations, contradictions and research gaps. | Add to team |
Feature Spec WriterTeam agent | Draft scope and observable acceptance criteria. | Accepted problem, constraints and explicit non-goals. | A reviewable feature brief and unresolved choices. | Add to team |
Prototype PlannerTeam agent | Plan the interaction needed to test an uncertainty. | A specific user task, uncertainty and available representation. | An interaction map and bounded prototype test brief. | Add to team |
QA Release ReviewerTeam agent | Assess release evidence with a clear verdict. | Exact change scope, acceptance cases and actual verification receipts. | A pass/fail/unverified matrix and ship-or-hold brief. | Add to team |
Set up your team in Finta
- Add each template individually. Use the Add to team action for each named coworker, review the import setup and complete its Getting Started skill.
- Give each role the right context. State the shared objective, then supply the authorized documents, notes or exports that role needs. Do not assume an unconnected system is available.
- Review the handoff. Accept or correct one return before giving its evidence and open questions to the next coworker. Importing four templates does not itself run a bulk team.
- Make recurrence deliberate. Complete a useful manual pass first. Review any supported routine and enable it only when the scope, inputs and permissions are right.
Keep the handoff useful
Begin with one shared objective and a current source pack. Give each coworker a distinct return, accept or correct that return, and carry its evidence and unanswered questions into the next step. Keep the approved artifact in the work context you actually use rather than relying on an agent’s memory of an earlier draft.
Planned research is not a finding, a prototype is not production software and a successful build is not complete release proof. The guides do not promise autonomous engineering execution.
Review the result
- Can the responsible person explain what changed and what decision is due?
- Does every important statement point to current evidence or a clearly labeled unknown?
- Are proposed actions, accepted commitments and completed work recorded as different states?
Sources and next steps
This playbook uses the current published Finta marketplace packages linked in the team table. Open each profile to review its instructions, context, Skills and any routines before importing.
- Customer Voice Researcher package and setup
- Feature Spec Writer package and setup
- Prototype Planner package and setup
- QA Release Reviewer package and setup
- Explore the product team guides
Browse the job guides
The eligible published guides for this department appear below. Each includes its own suggested team, setup inputs, worked artifact and copyable brief. Choose the closest job rather than importing every template at once.
Put your team to work
AI Product Agents: Turn Customer Feedback Into a Feature Brief
A traceable problem brief with contradictory feedback and acceptance criteria
AI agent team guideAI Release Review: Assemble a Clear Ship-or-Hold Brief
A release evidence matrix with unresolved gates and a human ship/hold decision
AI agent team guidePlan and Test a Product Prototype With an AI Team
A falsifiable prototype task, test plan and evidence-to-revision log