Limitless vs
SimpleFunctions.

SimpleFunctions is an API, CLI, and MCP server that sits above prediction markets as an agent and data layer. It aggregates live data from Kalshi and Polymarket, computes derived indicators, runs a causal thesis system, and provides autonomous trading via Portfolio Autopilot. Limitless is an on-chain trading venue that runs its own binary markets on Base L2. They serve different roles: Limitless is a market infrastructure layer; SimpleFunctions is the intelligence and automation layer built on top of established markets.

No. SimpleFunctions currently aggregates Kalshi and Polymarket data — it does not index Limitless's on-chain Base L2 markets. Limitless and the Kalshi/Polymarket universe are separate venues with different structures and settlement mechanisms. If you need exposure to Limitless's short-duration on-chain contracts specifically, you would access Limitless directly. SimpleFunctions covers 48K+ contracts across the two major regulated off-chain venues.

POST /api/thesis/create takes a plain-language claim and decomposes it into a causal tree of testable sub-claims, mapping each node to live contracts on Kalshi and Polymarket. An evaluation heartbeat then runs periodically — scanning news, refreshing prices, checking milestones, running an LLM evaluation, and updating confidence scores automatically. You can inject external signals at any time via /api/thesis/{id}/signal. Public theses are forkable by other users. No current competitor offers an equivalent system.

Portfolio Autopilot is SF's autonomous trading agent. It uses a 1M-context LLM and aggregates 13 data sources to form trade ideas. Before any execution, a 7-gate risk cascade runs sequentially — checking the kill switch, position limits, drawdown thresholds, regime classification, and more. The system is designed for traders who want supervised automation rather than manual order entry. It evaluates, gates, and executes without requiring per-trade confirmation, while keeping a full audit trail of gate decisions.

Yes. Running claude mcp add simplefunctions --url https://simplefunctions.dev/api/mcp/mcp adds 56 tools to Claude Code, Cursor, or any MCP-compatible client. Tools cover market search, orderbook queries, thesis creation and signal injection, trade ideas, calibration data, cross-venue pairs, and the full world snapshot. Limitless has no published MCP server. The SF MCP server requires no authentication for read-only tools.

GET /api/calibration returns live Brier scores for SF's own predictions, broken down by venue, category, and price bucket. Current published figures: Kalshi 0.20, Polymarket 0.12 on T-24h price over the past 90 days. This lets you verify SF's track record before building on it. Most competitors state accuracy claims in marketing copy without exposing underlying numbers. SF's calibration endpoint is publicly accessible — you can re-verify the figures yourself with curl at any time.

Yes. If you specifically need trustless on-chain settlement — no custodial counterparty, Pyth oracle resolution, Base L2 composability with other DeFi protocols — Limitless is the right tool. It is designed for on-chain participants who want binary prediction market exposure without relying on a centralised exchange. SimpleFunctions does not offer on-chain execution and does not index Limitless markets. The products address different threat models and different audiences.

Public REST endpoints, the MCP server, and the CLI do not require authentication for read operations — market prices, orderbook depth, computed indicators, calibration scores, and cross-venue arbitrage pairs are all openly accessible. Authenticated endpoints — thesis creation, intent execution, and autopilot — are free up to 15M tokens, then priced per token. There is no monthly subscription tier for data access.