Alternative · AI agent
Polyfactual vs
SimpleFunctions.
Polyfactual layers social-narrative truthfulness grading on top of prediction-market signals — a lens for narrative research and fact-checking. SimpleFunctions ships the agent layer above the same upstream venues: a causal-tree thesis system that decomposes any claim into testable sub-propositions with auto-evaluation cycles, an autonomous Portfolio Autopilot with a 7-gate risk cascade, computed indicators across 48K+ contracts, live Brier calibration scores, and a 56-tool MCP server. Different surfaces; different intended audiences.
Verified 2026-04 · public sources only · live SF data from /calibration
Verdict
Pick the one that fits how
you actually work.
Choose SimpleFunctions if
Choose SimpleFunctions if you are building agents, autonomous trading systems, or research pipelines that need calibrated probabilities with public Brier scores, a causal-tree thesis system with auto-evaluation cycles, regime classification across the full 48K-contract universe, computed indicators — implied yield, cliff risk index, liquidity availability score — and a 56-tool MCP server that integrates into Claude Code or Cursor in one command. The agent layer, not just the signal, is the product.
Choose Polyfactual if
Choose Polyfactual if your primary need is narrative-truthfulness grading — cross-referencing what prediction markets price against what social-narrative signals suggest about event likelihood. That intersection of market probability and narrative tracking is Polyfactual's stated product focus and serves a different audience than autonomous-trading builders or agent infrastructure consumers.
Same upstream prediction-market venues; Polyfactual adds narrative-truthfulness grading. SimpleFunctions ships the agent layer: world model, thesis system, indicators, autopilot, MCP.
At a glance
Three things that
actually differ.
Everything Polyfactual gives you — prediction-market signals and event probability tracking across venues — SimpleFunctions also gives you, on the same Kalshi and Polymarket feeds with 48K+ contracts normalised and indexed.
On top of that, SF ships a causal-tree thesis system, Portfolio Autopilot (1M-context LLM, 7-gate risk cascade), computed indicators (implied yield, cliff risk, liquidity availability score), and 56 MCP tools no current PM data product exposes.
SF also publishes live Brier scores at /api/calibration — Kalshi 0.20, Polymarket 0.12 on T-24h price, past 90 days — and exposes cross-venue arbitrage pairs at /api/public/cross-venue/pairs for programmatic verification.
Side by side
9 dimensions · verified 2026-04SimpleFunctionsKalshi + Polymarket normalised, 48K+ active contracts indexed at /api/public/markets.
PolyfactualPrediction-market signals across venues, blended with social-narrative truthfulness tracking as the stated core product.
SimpleFunctionsGET /api/public/market/{ticker}?depth=true returns a bid/ask ladder with spread and slippage estimate.
PolyfactualNot published in available public documentation.
SimpleFunctionsImplied yield, cliff risk index, liquidity availability score, event overround, τ-days, and regime label pre-computed across 48K contracts at /screen.
PolyfactualNarrative truthfulness scores layered over price signals; quantitative derived indicators are not documented publicly.
SimpleFunctionsSF models causal structure and propagates probability through testable sub-claims; social-narrative grading is not in scope.
PolyfactualCore capability — social-narrative tracking grades event truthfulness by cross-referencing market signals with narrative context.
SimpleFunctionsLive Brier scores by venue, category, and price bucket at /api/calibration — Kalshi 0.20, Polymarket 0.12 on T-24h price, 90-day rolling window.
PolyfactualNot published in available public documentation.
SimpleFunctionsPOST /api/thesis/create decomposes any sentence into a causal tree of sub-claims, scans Kalshi + Polymarket for edges, and runs an auto-evaluation heartbeat.
PolyfactualNot in scope; Polyfactual's framing centres on truthfulness grading rather than causal decomposition for trading.
SimpleFunctionsPortfolio Autopilot — 1M-context LLM, 13 data sources, 7-gate risk cascade including kill switch, position limits, drawdown gate, and regime check.
PolyfactualNot in scope.
SimpleFunctions56 tools via claude mcp add simplefunctions --url https://simplefunctions.dev/api/mcp/mcp — works with Claude Code, Cursor, and any MCP-compatible client.
PolyfactualNo MCP server published.
SimpleFunctionsPublic REST + MCP + CLI reads require no authentication. Authenticated thesis and intent execution is free up to 15M tokens, then pay-per-token.
PolyfactualNot publicly documented at the time of verification.
Methodology
Verified 2026-04 from public sources only — Polyfactual's documentation, public website, and publicly observable behaviour. We never claim non-public information about Polyfactual'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 an autonomous research agent that needs structured causal analysis of prediction-market events.
SimpleFunctions · best fit
SF's thesis system decomposes any research question into a causal tree, scans Kalshi and Polymarket for tradeable edges aligned with each node, and runs continuous auto-evaluation cycles. The 56-tool MCP server drops into Claude Code or Cursor in one command.
Polyfactual
Polyfactual's product centres on narrative-truthfulness grading rather than causal decomposition or autonomous agent tooling; it is not designed for this use case.
Scenario 02
Cross-referencing prediction-market prices with social-narrative signals to grade event truthfulness for journalism or fact-checking research.
SimpleFunctions
SF models causal structure and calibrated probability but social-narrative tracking and truthfulness grading are not part of SF's scope; this workflow would require supplementing SF data with a narrative-grading layer.
Polyfactual · best fit
This is Polyfactual's stated product focus — blending market signals with social-narrative tracking to produce truthfulness grades. They are purpose-built for this intersection.
Scenario 03
Integrating prediction-market data into a live AI trading agent via MCP.
SimpleFunctions · best fit
SF's 56-tool MCP server exposes normalised prices, computed indicators, thesis management, cross-venue arbitrage pairs, and Portfolio Autopilot controls — all accessible from Claude Code or any MCP client without authentication.
Polyfactual
Polyfactual does not publish an MCP server; API access details are not publicly documented.
Scenario 04
Auditing the accuracy of prediction-market prices against resolved outcomes programmatically.
SimpleFunctions · best fit
SF's /api/calibration endpoint returns live Brier scores segmented by venue, category, and price bucket — Kalshi 0.20, Polymarket 0.12 — updated over a rolling 90-day window and verifiable with a single curl call.
Polyfactual
Polyfactual does not publish calibration or Brier-score metrics in available public documentation.
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 Polyfactual and how is it different from SimpleFunctions?+
Polyfactual is an AI-powered platform that blends prediction-market signals with social-narrative tracking to grade event truthfulness. SimpleFunctions is an agent-first API, CLI, and MCP server built around a different set of primitives: a causal-tree thesis system, autonomous Portfolio Autopilot, computed indicators across 48K+ contracts, and live Brier calibration scores. Both draw on prediction-market data, but Polyfactual targets narrative research use cases while SimpleFunctions targets autonomous agent infrastructure and structured trading.
How does SimpleFunctions' causal thesis system work?+
POST /api/thesis/create takes a plain-English sentence and decomposes it into a causal tree of testable sub-claims. Each node is scanned against Kalshi and Polymarket for tradeable edges. A continuous evaluation heartbeat then runs for each thesis: news scan, price refresh, milestone check, LLM evaluation, and confidence update. You can inject external signals at any time via POST /api/thesis/{id}/signal, and public theses are forkable. No other prediction-market data product exposes this primitive.
What is Portfolio Autopilot in SimpleFunctions?+
Portfolio Autopilot is SF's autonomous trading agent. It uses a 1M-context LLM and 13 data sources to generate and evaluate position ideas, then passes each candidate through a 7-gate risk cascade — kill switch, position limits, drawdown gate, regime check, and additional guards — before any execution. The gates can be configured and the kill switch can halt all activity immediately. It is designed for researchers and builders who want autonomous PM exposure with explicit risk controls, not a manual trading UI.
Can I use SimpleFunctions with Claude Code or other AI agents?+
Yes. Run claude mcp add simplefunctions --url https://simplefunctions.dev/api/mcp/mcp to register SimpleFunctions as a 56-tool MCP server in Claude Code, Cursor, or any MCP-compatible client. Tools span market search, cross-venue arbitrage pairs, thesis management, indicator lookups, and Portfolio Autopilot controls. Read-only tools require no authentication.
How does SimpleFunctions publish its own calibration accuracy?+
The /api/calibration endpoint returns live Brier scores segmented by venue, category, and price bucket. Current figures: Kalshi 0.20, Polymarket 0.12 on T-24h prices, computed over a rolling 90-day window. The endpoint is public and unauthenticated — you can verify the numbers with a single curl call at any time. Most prediction-market data providers do not publish equivalent self-auditing metrics.
Does SimpleFunctions track narrative or social signals like Polyfactual?+
No. SimpleFunctions models causal structure — decomposing claims into testable sub-propositions and propagating probability through that tree — rather than tracking social-narrative signals or grading event truthfulness. If your primary need is narrative-truthfulness grading cross-referenced with market prices, Polyfactual is focused on that intersection. SF is built for autonomous agent pipelines, trading infrastructure, and structured research on the market itself.
Does Polyfactual have a public API?+
Polyfactual's public documentation does not detail a REST API with documented endpoints, authentication schemes, or rate limits at the time this page was verified. SimpleFunctions exposes a fully documented public REST API at /api/public, a CLI available via npm i -g @spfunctions/cli, and a 56-tool MCP server — all accessible without authentication for read operations.
What computed indicators does SimpleFunctions provide?+
SF pre-computes six indicators across all 48K+ active contracts: IY (implied yield — annualised return if the contract settles YES), CRI (cliff risk index — sensitivity to near-term binary resolution), LAS (liquidity availability score — depth-adjusted fill probability), EE (event overround — market inefficiency measure), τ-days (days to settlement), and regime label (adverse-selection classification). All are accessible via /screen without authentication.
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.