PRD Agent
Generates structured product requirements documents.
What it does
PRD writes a clear, structured product requirements document — problem, goals, scope, requirements, and success metrics — grounded in the evidence and decisions already on your canvas.
When to use it
Run it when you are ready to specify. It is strongest when your evidence, problem, and value prop are already defined so it can cite them.
A first draft that already cites the evidence
A PM who dreads the blank-page PRD
The situation. A problem is ranked and accepted, the value prop is mapped — but writing it all up into a coherent PRD is the part that always stalls.
What they do
- 1The PM runs the PRD agent from the Definition phase.
- 2It reads the problem cluster, evidence, and value prop already on the canvas.
- 3It produces a structured draft — problem, goals, scope, requirements, success metrics — citing the actual evidence.
- 4The PM edits inline, tightening scope and adding constraints.
The result. Instead of a blank page, they start from a grounded 80% draft and spend their time on judgment, not formatting. The PRD ships to review the same afternoon.
What it can do
- PRD generation
- Requirements engineering
- Success metrics
- Risk assessment
What you give it
When you run this agent you can tune it with these settings (sensible defaults are provided):
What you get back
- a written document you can edit and export
Every output lands on your canvas as a node you can edit, connect, and trace back to its source.
Examples
- Write PRD for notification center
- Create feature spec
- Document requirements
Lesson: a PRD is only as strong as its inputs
- What it is
- The PRD agent doesn’t invent the problem — it assembles a spec from what’s already defined on your canvas.
- Why it matters
- Run it on an empty canvas and you get generic boilerplate. Run it after synthesis and value-prop work, and it cites real evidence and real decisions — which is what makes a PRD persuasive.
Do the problem and value work first. The PRD agent rewards a well-prepared canvas.
Related agents
Blue Ocean Agent
Identifies uncontested market spaces using Blue Ocean Strategy.
SWOT Agent
Performs strategic SWOT analysis — strengths, weaknesses, opportunities, and threats.
Cognitive Bias Auditor
Audits your design for cognitive biases and suggests debiasing strategies.
UX Laws Agent
Applies psychological principles and UX laws to design.