agentleFS
ratacatratacat/ratacats-skills/skills/kalshi-prediction-market/SKILL.md

kalshi-prediction-market

Skill9 starsChanged 5 months agoMITView on GitHub

Context and working knowledge for the Kalshi prediction-market domain. Use this skill whenever the user asks about Kalshi markets, tickers, order books, pricing, settlement, or the Kalshi API/WebSocket.

The file

We split this skill into the rules it gives an agent

7 rules. About 0 are common, 7 are distinctive, and 0 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈7

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 0

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Every tradable contract is Yes/No on a real‑world outcome.

    line 20
  2. Distinctive:

    If a Yes contract trades at $0.74, the market implies ~74% chance of Yes.

    line 25
  3. Distinctive:

    No price is complementary (roughly 1 − Yes, ignoring fees/spread).

    line 26
  4. Distinctive:

    When you see “market,” clarify whether it means an event page (container) or a specific market outcome (binary leg).

    line 40
  5. Distinctive:

    When interpreting Kalshi market JSON:

    line 44
  6. Distinctive:

    When asked “how does this resolve?”, reference the series rules and settlement source, then restate in plain language.

    line 73
  7. Distinctive:

    Explain both sides (Yes/No) and spreads: when discussing pricing or order books.

    line 107

≈ “Common” is an estimate: at least 1 in 5 of the 372,429 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Kalshi Prediction Market

Kalshi is a regulated prediction-market exchange. This skill provides the domain model, trading mechanics, and API conventions you need to reason about Kalshi data and to explain it clearly to users.

Core Mental Model

  1. Binary event contracts
    • Every tradable contract is Yes/No on a real‑world outcome.
    • A winning side pays $1, losing side pays $0.
    • Prices between $0.01–$0.99 represent implied probability.
  2. Implied probability
    • If a Yes contract trades at $0.74, the market implies ~74% chance of Yes.
    • No price is complementary (roughly 1 − Yes, ignoring fees/spread).
  3. Fully collateralized
    • Users pay maximum loss up‑front. No margin/leverage.
    • You can never lose more than you spend on contracts.

Data Hierarchy

Kalshi uses a strict hierarchy:

  • Series → template for recurring markets (shared rules/settlement).
  • Event → specific instance within a series (a real‑world occurrence).
  • Market → single binary contract within an event (one Yes/No outcome).

When you see “market,” clarify whether it means an event page (container) or a specific market outcome (binary leg).

Market Objects: What Fields Mean

When interpreting Kalshi market JSON:

  • ticker: unique identifier (string).
  • event_ticker / series_ticker: parent identifiers.
  • title / subtitle: human‑readable question and clarification.
  • yes_bid / yes_ask (cents) and _dollars: best prices to buy/sell Yes.
  • no_bid / no_ask: best prices to buy/sell No.
  • last_price: last traded Yes price.
  • volume / volume_24h / open_interest: activity and outstanding contracts.
  • open_time / close_time / expiration_time: lifecycle timestamps.
  • status: initialized, active/open, closed, settled.
  • result / settlement_value: set after resolution.

Trading Mechanics to Explain

  • Order book on both Yes and No sides.
  • Quick/market order crosses current spread for immediate fill.
  • Limit order rests at a chosen price; may add liquidity.
  • Closing a position = taking the opposite side later (sell Yes or buy No).
  • Mutually exclusive events contain multiple markets where at most one can settle Yes.

Fees on Kalshi are variable/quadratic, roughly a percent of potential profit; maker orders may be discounted.

Settlement & Resolution

  • Each series defines official settlement sources and rules.
  • Markets usually close before the strike/decision time, then settle after confirmation.
  • Some markets can resolve early if can_close_early is true.

When asked “how does this resolve?”, reference the series rules and settlement source, then restate in plain language.

API Conventions You Should Use

Public data (no auth needed):

  • GET /series
  • GET /events (events include their markets)
  • GET /markets
  • GET /market/{ticker}
  • GET /market/orderbook
  • GET /market/candlesticks
  • GET /market/trades
  • GET /exchange/status

Trading/account (auth required):

  • POST /orders, DELETE /orders/{id}, GET /orders/{id}
  • POST /order-groups and related order‑group endpoints
  • GET /portfolio/balance, GET /portfolio/positions, GET /portfolio/fills

Auth uses an API key id plus RSA signature headers:

  • KALSHI-ACCESS-KEY
  • KALSHI-ACCESS-TIMESTAMP
  • KALSHI-ACCESS-SIGNATURE

Real‑time updates arrive via WebSocket subscriptions to tickers.

How to Apply This Skill When Answering

  1. Identify the exact event or market the user means if they’re vague.
  2. Always distinguish Series/Event/Market and restate which level you’re discussing.
  3. Convert price to probability explicitly when helpful.
  4. Explain both sides (Yes/No) and spreads when discussing pricing or order books.
  5. Cite rules + settlement source for resolution questions.
  6. Stay neutral: describe mechanics and risks; don’t give financial advice.

Examples

  • “This is a Kalshi market ticker — a binary contract paying $1 if Yes. At $0.62, the market implies ~62% Yes probability.”
  • “The event is mutually exclusive, so each candidate outcome is a separate market. Exactly one can settle Yes.”
  • “To get real‑time prices, subscribe to the market tickers on the Kalshi WebSocket.”

For more detail, see reference.md.

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.