# KalshiBots — Algorithmic Trading Systems for Kalshi Prediction Markets

> KalshiBots is a marketplace of 15 pre-built algorithmic trading systems ("bots") for Kalshi, a regulated U.S. prediction-market exchange. Each bot runs a distinct decision strategy (adversarial/contextual bandits, LLM-driven news research, ensemble/momentum/mean-reversion, or continual self-learning). Licensing is available on request through the site's request form; no public list price is published.

## What this site is

- A catalog/dashboard ("The Collection") of 15 trading systems, each with a tag (TX-01 through TX-15), a strategy description, a feature list, and a "Request" button.
- Every "Request" button opens a request form (name, work email, optional requirements) submitted to our own Cloudflare Pages Function, which verifies a Cloudflare Turnstile challenge and emails the team server-side via Resend. The destination address is never present in the page source, and no third-party form processor is involved. A direct email fallback link is also shown below the form in case the request ever fails to send.
- The site's stats banner (system count, algorithms used, architecture, auth scheme) describes the catalog itself. No backtested returns, APR, or win-rates are published anywhere on the site — real trading performance depends on the account, timing, and risk settings used.

## How to request a bot

1. Open https://www.kalshibots.net/ and go to "The Collection" section (`#catalog`).
2. Click "Request" on the desired bot's card (grid view) or row (table view).
3. Submit the request form. A member of the team follows up by email with pricing and deployment details.
4. There is no public price list — all pricing is "Price Upon Request" / negotiated by license tier.

## What you need to run a bot

- A Kalshi account (demo or live) with API credentials.
- A computer or server that stays on to run the bot continuously.
- For AI/LLM-driven bots specifically: an API key for the underlying language model.
- The exact setup checklist is sent per-bot after a request is submitted — it is not published on the site.

## How the bots work (plain-language)

Each bot runs on two loops: a fast "trading loop" that scans Kalshi's prediction markets and places trades using its strategy (momentum, news analysis, statistical models, etc.), and a slower "learning loop" that periodically reviews trade performance and adjusts the bot's strategy weights and risk settings — so it improves over time without manual reprogramming.

## Bot Catalog (15 systems)

| Tag | Name | Category | Key Features |
|---|---|---|---|
| TX-01 | The Apex Intelligence | AI / Self-Learning | Continual self-learning, AI market reading, unified risk management, market making |
| TX-02 | The News Reader | AI / Self-Learning | Live web search integration, LLM-based news/event analysis, adaptive multi-strategy blend |
| TX-03 | The Safe Guardian | Safe / Capital Guardian | Strict risk manager, contextual-bandit category selection, steady-growth focus |
| TX-04 | The Speed Demon | High Speed / Asynchronous | Asynchronous execution, automatic take-profit, automatic stop-loss |
| TX-05 | The Strategist | AI / Self-Learning | Dynamic strategy switching, adaptive thresholds, machine-learning core |
| TX-06 | The Data Cruncher | AI / Self-Learning | Live ML model training, data-driven predictions, smart capital allocation |
| TX-07 | The Reliable Classic | Safe / Capital Guardian | Proven rule-based trading, momentum tracking, undervalued-opportunity finding |
| TX-08 | The Platform Hopper | Safe / Capital Guardian | Full-stack dashboard, market-category learning, cross-platform (multi-exchange) ready |
| TX-09 | The Cautious Explorer | Safe / Capital Guardian | Cryptographically signed secure trading, balanced strategy mix, tight exposure caps |
| TX-10 | The Adversarial Strategist | AI / Self-Learning | Adversarial-bandit strategy core, multi-strategy mixing, paper-trading validated |
| TX-11 | The Confidence Optimizer | AI / Self-Learning | Contextual-bandit category selection, automated self-tuning (hyperparameter search), resilient to API errors |
| TX-12 | The Value Hunter | Safe / Capital Guardian | Momentum + value ensemble, disciplined (Kelly-style) position sizing, risk-adjusted strategy selection |
| TX-13 | The Analyst | AI / Self-Learning | Momentum + news ensemble, win-rate-driven reweighting, built-in performance analytics |
| TX-14 | The Async Operator | High Speed / Asynchronous | Fully asynchronous engine, self-improving bandit core, active take-profit/stop-loss management |
| TX-15 | The Portfolio Balancer | Safe / Capital Guardian | Diversified momentum/mean-reversion/value signal blend, automated risk tuning, graceful market-feed fallback |

## Categories

- **AI / Self-Learning** — bandit algorithms, LLM research, or model retraining drive decisions.
- **Safe / Capital Guardian** — conservative, risk-first bots that prioritize capital preservation over yield.
- **High Speed / Asynchronous** — fully async execution engines built for fast fills and active position management.

## Security & trust notes

- The request form uses a honeypot field and Cloudflare Turnstile (verified server-side) to block automated spam submissions.
- No student, developer, or third-party names are published on this site — bot identifiers are anonymized product tags (TX-01…TX-15), not personal identifiers.
- Trading prediction markets involves substantial risk; past or simulated performance does not guarantee future returns.

## Contact

- Use the "Request" form on the site for licensing/pricing inquiries — this is the only supported contact channel.

## Site Pages

- index.html — the entire site (single-page marketplace + request-form modal). There are no other pages.