Alternative · AI agent
Polybro vs
SimpleFunctions.
Both are autonomous prediction market agents, but they operate at different levels of the stack. Polybro specialises in Polymarket signal generation via structured research — academic papers, news feeds, and live market data. SimpleFunctions ships the full agent layer above raw signals: a causal-tree thesis system with auto-evaluation heartbeats, Portfolio Autopilot with a 7-gate risk cascade, computed indicators across 48K+ contracts on both Kalshi and Polymarket, and a 56-tool MCP server that integrates directly with Claude Code or Cursor.
Verified 2026-04 · public sources only · live SF data from /calibration
Verdict
Pick the one that fits how
you actually work.
Choose SimpleFunctions if
You need the full agent layer — structured thesis decomposition into causal sub-claims with auto-evaluation cycles, Portfolio Autopilot backed by a 1M-context LLM and 7-gate risk cascade, computed indicators (implied yield, cliff risk, liquidity availability score, event overround) pre-screened across 48K+ contracts on both Kalshi and Polymarket, live Brier scores you can audit at /api/calibration, and 56 MCP tools that drop into Claude Code or Cursor in one line.
Choose Polybro if
You want a Polymarket-focused autonomous agent whose signal generation is specifically grounded in academic literature alongside news and live data. Polybro is purpose-built for that research-intensive workflow, and its product targets traders who prioritise structured literature-informed signals over a general-purpose cross-venue world model.
Polybro: research-intensive Polymarket signal generation. SimpleFunctions: full agent layer — thesis system, autopilot, indicators, MCP — spanning Kalshi and Polymarket.
At a glance
Three things that
actually differ.
Everything Polybro gives you — autonomous signal generation from live data and news feeds on Polymarket — SimpleFunctions also gives you, with the same Polymarket coverage extended to Kalshi across 48K+ normalised contracts.
On top of that, SF ships a causal-tree thesis system with rolling auto-evaluation, Portfolio Autopilot (1M-context LLM, 7-gate risk cascade), 56 MCP tools, and computed indicators that no current PM signal product exposes.
SF also publishes live Brier scores at /api/calibration — Kalshi 0.20, Polymarket 0.12 on T-24h price, past 90 days — letting you audit SF's own accuracy in a way Polybro does not.
Side by side
9 dimensions · verified 2026-04SimpleFunctionsKalshi + Polymarket normalised, 48K+ active contracts indexed across both venues.
PolybroPolymarket only, per public product description.
SimpleFunctionsTrade ideas at /api/public/ideas with conviction + catalyst; thesis system scans both venues for tradeable edges on each causal sub-claim.
PolybroStructured research across academic papers, news feeds, and live Polymarket data to generate directional signals.
SimpleFunctionsGET /api/public/market/{ticker}?depth=true returns the full bid/ask ladder, spread, and slippage estimate.
PolybroNot documented publicly.
SimpleFunctionsImplied yield, cliff risk index, liquidity availability score, event overround, τ-days, and regime label pre-computed across 48K contracts at /screen.
PolybroNot published; signal output is directional, not decomposed into named sub-indicators.
SimpleFunctionsLive Brier scores at /api/calibration — broken down by venue, category, and price bucket, past 90 days.
PolybroNot published.
SimpleFunctionsPOST /api/thesis/create decomposes any natural-language thesis into a causal tree, propagates probabilities, and runs a news-scan → LLM-eval heartbeat automatically.
PolybroNot in scope; Polybro generates signals but does not expose a user-defined causal decomposition API.
SimpleFunctionsPortfolio Autopilot — 1M-context LLM, 13 data sources, 7-gate risk cascade (kill switch, position limits, drawdown gate, regime check) before any execution.
PolybroAutonomous Polymarket trading agent with research-driven signal generation as its described execution model.
SimpleFunctions56 tools via claude mcp add simplefunctions --url https://simplefunctions.dev/api/mcp/mcp; works with Claude Code, Cursor, any MCP client.
PolybroNo MCP server published.
SimpleFunctionsPublic REST + MCP + CLI reads require no authentication; pay-per-token only on thesis and intent execution, free up to 15M tokens.
PolybroNot publicly documented.
Methodology
Verified 2026-04 from public sources only — Polybro's documentation, public website, and publicly observable behaviour. We never claim non-public information about Polybro's internals. SimpleFunctions claims on this page are computed live from /api/calibration, /api/public/cross-venue/pairs, and /api/public/markets — you can re-verify them yourself with curl.
Use cases
Same data, different
best fit per scenario.
Scenario 01
Building a multi-venue AI agent that tracks geopolitical risk across Kalshi and Polymarket simultaneously.
SimpleFunctions · best fit
SF's thesis system decomposes the thesis into sub-claims, maps each to live contracts on both Kalshi and Polymarket, and re-evaluates on a rolling heartbeat. The world snapshot at /api/agent/world provides a single ~800-token context injection for any LLM agent. Portfolio Autopilot can execute positions once confidence thresholds are met.
Polybro
Polybro's product is scoped to Polymarket, so cross-venue coverage is outside its described capability. For single-venue Polymarket research, Polybro's academic and news signal pipeline could complement a bespoke agent build.
Scenario 02
Running an autonomous Polymarket agent whose signals are grounded in academic literature and structured news analysis.
SimpleFunctions
SF provides trade ideas with conviction scores at /api/public/ideas and thesis auto-evaluation via news scan and LLM review, but its research pipeline does not specifically index academic papers — that capability is not in scope.
Polybro · best fit
Polybro is purpose-built for this use case — autonomous signal generation via structured research across academic papers, news feeds, and live Polymarket data. If academic-literature grounding is a hard requirement for your Polymarket strategy, Polybro's product is built around exactly that workflow.
Scenario 03
Decomposing a complex macroeconomic thesis into tradeable sub-claims and surfacing relevant prediction market contracts.
SimpleFunctions · best fit
POST /api/thesis/create accepts the thesis sentence, builds a causal sub-claim tree, and surfaces Kalshi/Polymarket contracts that proxy each node; the evaluation heartbeat refreshes as macro data releases. No other PM data product currently exposes a structured causal decomposition API.
Polybro
Polybro generates directional signals from research but does not expose a causal decomposition API for user-defined theses — you receive a signal, not a forkable causal tree.
Scenario 04
Embedding prediction market intelligence into a Claude Code or Cursor workflow via MCP.
SimpleFunctions · best fit
One command — claude mcp add simplefunctions — connects 56 tools to your LLM coding context: query markets, pull computed indicators, check calibration, run searches, and trigger trade ideas. No API key needed for reads.
Polybro
Polybro does not publish an MCP server; integrating it into an LLM IDE would require building a custom connector on top of whatever API surface Polybro exposes, if any.
Live data
The SimpleFunctions claims on this page are not marketing copy. Brier scores, market counts, and cross-venue pair counts are computed live from /calibration, /screen, and /api/public/cross-venue/pairs. All public, all free, all CC-BY-4.0.
FAQ
What is Polybro and what does it do?+
Polybro is an autonomous AI trading agent focused on Polymarket. It conducts structured research across academic papers, news feeds, and live market data to generate trading signals. Its product is oriented toward Polymarket-only traders who want directional signals grounded in literature and news research, rather than a generalised cross-venue world model or a structured thesis decomposition API.
How does SimpleFunctions's thesis system work?+
POST /api/thesis/create accepts any natural-language thesis sentence. The system decomposes it into a causal tree of testable sub-claims, assigns propagated probabilities, and scans both Kalshi and Polymarket for contracts that represent tradeable edges on each node. A heartbeat cycle then runs continuously — news scan, price refresh, milestone check, LLM evaluation, confidence update. You can inject external signals at any time via /api/thesis/{id}/signal. Public theses are forkable by other users. No competitor currently exposes this workflow.
Does SimpleFunctions have autonomous trading built in?+
Yes. Portfolio Autopilot uses a 1M-context LLM, 13 data sources, and a 7-gate risk cascade — including a kill switch, position limits, drawdown gate, and regime classification — before any order is placed. It operates on SF's calibrated world model rather than raw price feeds, and is designed to abort at any gate rather than proceed on ambiguous signal. This is not a simple signal-follower; it is a risk-managed execution layer.
Does Polybro cover Kalshi?+
Based on public product descriptions, Polybro is a Polymarket-focused product. SimpleFunctions covers both Kalshi and Polymarket with normalised prices across 48K+ active contracts, cross-venue matched pairs for arbitrage detection at /api/public/cross-venue/pairs, and indicators computed uniformly across both venues. If your strategy requires Kalshi coverage alongside Polymarket, SF is the relevant tool.
Can I use SimpleFunctions with Claude Code or Cursor?+
Yes. SimpleFunctions runs a 56-tool MCP server. Add it with: claude mcp add simplefunctions --url https://simplefunctions.dev/api/mcp/mcp. Once connected, your Claude Code or Cursor session can query markets, run scans, evaluate theses, pull calibration data, retrieve indicators, and trigger trade ideas directly from the assistant interface. No API key is required for read operations.
How do I verify SimpleFunctions's accuracy claims?+
GET /api/calibration returns SF's live Brier scores broken down by venue, category, and price bucket, computed over the past 90 days. Current readings: Kalshi 0.20, Polymarket 0.12 on T-24h price. The endpoint is public and unauthenticated — you can run it yourself with curl and compare against any other source. Polybro does not publish equivalent calibration data based on publicly available information.
What does 'agent-first' mean in SimpleFunctions's design?+
Agent-first means the primary API contract is designed for LLM agents and autonomous systems, not human dashboards. The world snapshot at /api/agent/world returns ~800 tokens of structured market context for direct injection into LLM context windows. The MCP server, thesis auto-evaluation heartbeat, and delta endpoint exist so agents can subscribe to change streams rather than polling every field on every tick. Human interfaces are built on the same API surface, not a separate one.
Start for free.
Public endpoints are free for normal usage and rate-limited for reliability. Authenticated endpoints are free up to 15M tokens, then pay per token. No credit card to start.