Shadow Board

Inspiration

Most AI tools are built to help you move faster. I wanted to build one that helps you slow down before a risky decision. People often make choices with weak evidence, hidden fear, or too much agreement around them. Shadow Board was built to challenge that pattern. The idea is simple: Before you act, put your decision on trial.

What it does

Shadow Board helps users stress-test important decisions before they commit. A user enters:

• The decision they are considering
• Why it matters now
• Their current leaning
• The stakes
• The deadline
• The evidence they already have
• What they may be afraid to admit Shadow Board then reviews the decision through six opposing roles:
• Advocate
• Skeptic
• User
• Operator
• Future You
• Judge The final report gives the user:
• Go, Revise, Wait, or Kill verdict
• Confidence score
• Decision scorecard
• Top risks
• Missing facts
• Future failure letter
• Kill criteria
• 48-hour evidence test
• Next actions
• Saved and shareable report history

How I built it

I built Shadow Board as a solo developer using MeDo. I used multi-turn prompting to shape the app step by step. First, I built the core landing page and decision intake flow. Then I refined the product into a structured decision-review system with six role cards, a verdict, scorecard, risks, missing facts, and action plan. After that, I added saved decision history, shareable reports, demo mode, and an About page explaining how MeDo was used. MeDo helped generate the app structure, UI, routing, report layout, saved records, and public deployment.

Challenges I ran into

The hardest part was making Shadow Board feel like a product, not a generic AI form. A simple prompt box would not be enough. The app needed a clear flow:

• Ask the right questions
• Challenge the user from different angles
• Give a fast verdict
• End with a concrete test Another challenge was report design. Long AI answers are easy to ignore. I had to shape the output into cards, scores, verdicts, risks, and next actions so users could scan it quickly. The history page also needed cleanup so saved decisions looked polished and useful.

Accomplishments that I'm proud of

I am proud that Shadow Board has a clear point of view. It does not just give advice. It challenges the user. The strongest part is the final report. It turns a messy decision into a clear brief with a verdict, risks, missing facts, kill criteria, and a 48-hour evidence test. I am also proud of the Future You section. It shows how the decision could fail six months later, which makes the review more direct and memorable. As a solo developer, I was able to go from idea to working deployed app quickly by using MeDo as the build partner.

What I learned

I learned that no-code AI tools still need strong product direction. MeDo can generate a working app fast, but the final quality depends on how clearly the workflow is shaped. The biggest improvement came from narrowing the app. Instead of building a broad decision assistant, I focused Shadow Board on one job: Help users find what could make them wrong before they act. That focus made the app easier to explain, easier to demo, and more useful.

What's next for Shadow Board

Next, I want to add:

• Team decision reviews
• Outcome tracking
• Decision follow-ups after 7, 30, and 90 days
• Comparison between predicted risks and real results
• Templates for founders, product teams, career moves, and major purchases
• Exportable decision briefs for teams The long-term goal is to make Shadow Board a decision journal that helps people track not only what they chose, but why they chose it and what happened after.

Built With

• ai-generated-role-reviews
• appmedo-deployment
• medo
• public
• react
• reports
• saved-report-history
• shareable
• supabase
• tailwind-css
• typescript