Morning Desk

Morning Desk is a Base-built pocket analyst that turns news and Polymarket data into a scrollable briefing feed, with cards showing odds, DD bullets, tickers, and on-chain user sentiment.

Comment

Inspiration

Morning finance desks in investment / commercial banking and S&T start the day by scanning headlines, research, and market moves, then turning that firehose into a focused view of risk and opportunity. I wanted a friendly version of that: a pocket analyst that turns chaotic news and prediction markets into a clean, actionable morning brief.

A big part of the motivation was accessibility, taking the rigor of institutional finance (probabilities, scenarios, thesis vs. risk) and packaging it so an everyday person can understand why markets are moving, not just that they are. Instead of needing a Bloomberg terminal, you get an interface that explains “here’s what the market thinks, here’s why, here’s what could go wrong.”

What it does

Morning Desk is a Base-built “pocket analyst.” It pulls curated Polymarket markets and macro themes into a scrollable feed of briefing cards. Each card shows:

  • Implied probabilities (e.g. (P(\text{event}) \approx 0.52))
  • Analyst-style due-diligence bullets (facts, drivers, risks, catalysts)
  • Related tickers (stocks, ETFs, crypto)
  • Source links for deeper reading

Users can:

  • Filter by topic (US Politics, Crypto, Energy, Trade)
  • Ignore noise they don’t care about
  • Mark what they agree or disagree with
  • Track cards into a Portfolio tab, which aggregates their bullish / bearish / watchlist themes into a simple “strategy snapshot.”

Under the hood, a Base smart contract stores per-topic sentiment for each wallet as the first step toward on-chain view profiles and future banker-bots.

How we built it

I built Morning Desk as a three-layer system:

  1. Smart contract (Solidity + Hardhat) on Base Sepolia

    • ViewProfile contract stores topic views (Bullish / Neutral / Bearish) per address.
    • Uses bytes32 topic IDs (e.g. US_POLITICS, CRYPTO) for flexibility.
    • Includes events, custom errors, and tests for all main paths.

  2. Backend router (Node.js + Express + TypeScript)

    • Serves a /feed API of BriefingItems seeded with realistic, evergreen mock data for US politics, crypto, and policy topics like SMRs and softwood lumber.
    • Each item includes implied probability, DD bullets, related tickers, and source links.
    • Designed to be easily swapped to real Polymarket + news APIs later.

  3. Frontend (Next.js + React + TypeScript)

    • Calls the API, renders the feed, and integrates wagmi/RainbowKit for Base wallet connection.
    • UI includes topic filters, Polymarket previews, due-diligence bullets, sources, and local Agree / Disagree / Track / Ignore interactions.
    • A Portfolio tab turns tracked cards + opinions into a simple “strategy snapshot” so non-experts can see their bullish, bearish, and watchlist themes at a glance.

Challenges we ran into

  • Scope vs. time: I had to balance ambition with hackathon time, wiring contracts, backend, and frontend cleanly without getting stuck on infra or deployment.
  • Polymarket + TOS safety: Designing something Polymarket-aligned but TOS-safe meant avoiding custodial trading, bots, or geo-bypass and focusing on tooling and analysis, not execution.
  • Mock data that still feels real: I wanted the content to feel current and analytical while using mock data only, so I iterated a lot on how the bullets, probabilities, and sources are written. The goal was to feel like a real junior-analyst note, not lorem ipsum.
  • Explaining finance without jargon: Making serious concepts (implied probabilities, macro drivers, risk scenarios) understandable without dumbing them down was a constant tension, especially while sleep-deprived.

Accomplishments that we’re proud of

  • End-to-end product, not just parts: Morning Desk is a full flow, not just a Figma or a single contract. The contracts compile and pass tests, the API serves structured briefing data, and the Next.js UI feels like a real research product.
  • Analyst-style cards: The cards combine Polymarket odds, analyst-style DD, related assets, and sources so you can understand a thesis in seconds, the way a desk analyst would summarize it.
  • View profiles as rails for future agents: The on-chain ViewProfile contract gives a clear path for future agents or banker-bots to read a user’s views and generate structured trade “recipes” on top.
  • Accessibility of rigor: I’m proud that Morning Desk tries to bring the rigor of institutional decision-making, clear theses, probabilities, risks, into a UI that feels approachable to someone who doesn’t work in finance.

What we learned

  • How to think beyond “just another DeFi app” and design for fast, fun, value-in-20-seconds UX, which is what Base emphasized in the kickoff.
  • Hands-on experience with Base tooling: Hardhat on Base Sepolia, wagmi/RainbowKit, and how to architect around a Layer 2.
  • A new way to see prediction markets: not just as places to trade, but as a data layer that encodes crowd (P(\text{event})) and can drive research, dashboards, and agents.
  • The power of mock-first development: by stubbing data and future LLM calls, I could still ship a polished experience on a short timeline instead of getting blocked on external APIs.
  • How important copy is: small changes in the DD bullets and labels have a huge effect on whether non-experts feel like the tool is “for them.”

What’s next for Morning Desk

  • Real integrations: Plug in real Polymarket APIs and news sources, then swap the template bullets for true LLM-generated due diligence (with clear labeling).
  • On-chain opinions: Wire Agree / Disagree / Track to the Base ViewProfile contract so agents can read a user’s on-chain views and generate basket trade “recipes” safely (no custody, just plans).
  • Deeper strategy modeling: Evolve the Portfolio “strategy snapshot” into richer theme and risk analytics, still in plain language, so everyday users can see how their views cluster across politics, crypto, and policy.
  • Mini-app + more regions: Explore a Base mini-app version, richer topic coverage (including more Canadian policy), and institution-facing analytics for funds or research desks using aggregated, privacy-respecting on-chain sentiment.

Overall, Morning Desk is an experiment in taking the serious, math-driven side of finance and making it understandable and usable for normal people, without losing the rigor.

Built With

Share this project:

Updates