Library by layer

Score every market across volume, spread, depth, dispute history, rule clarity, implied volatility, and time-to-resolution into a single quality rank. Sits before every strategy and replaces ad-hoc per-strategy quality checks with a shared, consistent score.

0.2 Discovery PLANNED

MarketQualityRanker

Map every Polymarket market to a known real-world calendar event (elections, sports fixtures, court dates, FOMC releases, earnings, debates) and emit a time-tagged ObservationReport so strategies can act on pre-event positioning logic.

0.3 Discovery PLANNED

EventCalendarMapper

Maintain a ranked queue of the best candidate markets per registered strategy type by combining MarketQualityRanker scores, EventCalendarMapper proximity signals, and per-strategy fit-scores into a single ordered list that strategies consume without re-running discovery.

0.4 Discovery PLANNED

OpportunityQueue

Poll the Gamma API at high frequency to detect newly listed markets as soon as they appear, parse their initial metadata, and emit ObservationReports so strategies can evaluate early liquidity opportunities before the book fills.

0.5 Discovery PLANNED

NewMarketWatcher

Detect semantically identical or dangerously overlapping Polymarket markets to prevent accidental correlated exposure and to surface cross-market arbitrage opportunities. Emits ObservationReports tagging each duplicate cluster with a similarity score.

0.6 Discovery PLANNED

DuplicateMarketDetector

1.1 Risk LIVE v3.5 · wired

Risk

19 modules · 7 live · 1 beta · 11 planned

The authority layer. Sixteen guardrails vote on every OrderIntent before it leaves the process — account limits, kill switches, oracle risk, liquidity depth, rate caps, inventory unwinds, blacklists, jurisdiction gates, and guards covering user suitability, capital allocation, correlation shocks, model drift, tail-loss, fee economics, settlement exposure, and override discipline. Each runs in its own goroutine and short-circuits on REJECT — there is no retry-with-different-params loophole.

PortfolioGuard reference

PortfolioGuard enforces account-wide exposure limits across every running strategy simultaneously. It tracks aggregate notional, rolling 24-hour drawdown, per-market concentration, and correlated-cluster concentration. When an incoming order would breach any of these limits, it is either downsized to the safe remaining budget or rejected outright. PortfolioGuard does not change the strategy intent, the market, or the direction — it only sets the maximum size an order is allowed to carry.

1.2 Risk LIVE v3.5 · wired

KillSwitch reference

KillSwitch is the top-level emergency stop for the entire trading system. It can be triggered automatically when intraday or weekly drawdown exceeds a threshold, when the order-reject rate spikes above a circuit-breaker level, or when a market data feed is lost with open positions. It can also be triggered manually via the Admin UI. Once active, KillSwitch rejects every incoming OrderIntent without exception until a manual reset is performed (if require_manual_reset=true) or the trigger condition clears. It does not modify orders — it only blocks them entirely.

OracleRiskMonitor reference

1.3 Risk LIVE

OracleRiskMonitor watches the UMA Optimistic Oracle queue for proposals and active disputes on markets where open positions exist. When a market enters a resolution proposal or a dispute window, the bot can block new orders on that market, require reduced size, or flag a position for review. It protects against the scenario where a trade is submitted into a market moments before a contested resolution flips the outcome. It never overrides the strategy intent or changes the direction of an order — it only controls whether and how much the order is permitted to proceed.

LiquidityGuard prevents strategies from placing orders that would consume too much of the visible order-book depth on a given market. It checks book depth, spread, and top-of-book freshness on every OrderIntent and either approves, downsizes, or rejects the order. It cannot change the market, the direction, or the strategy intent — only the size and the timing of execution.

1.4 Risk LIVE

LiquidityGuard reference

RateLimitGovernor prevents the system from exceeding Polymarket's CLOB order send-rate limits at both the per-market and per-account levels. It reads live rate-limit headers from CLOB responses (X-RateLimit-Remaining, X-RateLimit-Reset), maintains sliding-window counters, and either delays (reshapes) or rejects OrderIntents when the remaining budget falls below configured thresholds. Cancel and risk-flatten requests always receive priority allocation over new open orders. The bot is fail-closed: if rate-limit state cannot be determined, new order intents are rejected until state is re-established.

1.5 Risk LIVE

RateLimitGovernor

InventoryUnwinder detects when a position has breached its concentration or capital limit — either because an OrderIntent would push it over, or because an existing position is already over the limit (e.g. after a parameter change or strategy crash). When a breach is detected it generates unwind OrderIntents targeting the source bot, using the NegRiskAdapter on Polygon for negRisk markets, and routes them back into the execution pipeline. It can also hard-reject incoming intents that would worsen an already-breached position. Builder codes from the original strategy are preserved on unwind intents for attribution. Fail-closed: if position data is unavailable, all new intents for the affected market are rejected.

1.6 Risk LIVE

InventoryUnwinder

BlacklistKeeper maintains a dual-registry of banned market condition IDs and banned counterparty wallet addresses. On every OrderIntent it checks both registries and hard-rejects any intent whose target market or counterparty appears on either list. The bot also monitors for ambiguity signals (undefined resolution sources, prior dispute history, time-to-resolution below threshold) and rejects structurally hostile markets before execution. It is fail-closed: if the registry cannot be read, the intent is rejected.

1.7 Risk BETA

BlacklistKeeper

Guardrail Limited live

1.8 Risk LIVE v3.5 · wired

ComplianceGate

ComplianceGate enforces Polymarket's terms-of-service access policy on every OrderIntent before it reaches execution. It validates (1) that the originating wallet has completed Polymarket onboarding, (2) that the user's jurisdiction is not on the blocked list, (3) that the wallet address does not appear on OFAC or other configured sanctions lists, and (4) that the target market is eligible for trading (not restricted by category policy). When any check fails the order is rejected outright; for geopolitical markets the bot applies NegRisk-aware category rules before passing. ComplianceGate is fail-closed: any inability to verify policy results in a hard reject.

StrategySuitabilityGate screens every OrderIntent against the user's declared experience tier and capital envelope, blocking strategy types that exceed the user's configured risk profile. It prevents novice-tier users from executing advanced multi-leg or high-leverage strategies without explicit elevation.

1.9 Risk PLANNED

StrategySuitabilityGate

CapitalAllocator carves the user's total risk budget across active strategies and refuses new orders that would cause any strategy slice or the aggregate portfolio to exceed its configured allocation. It emits a RESHAPE constraint when a downsize is possible, and HARD_REJECT when the budget is exhausted.

1.10 Risk PLANNED

CapitalAllocator

CorrelationShockGuard monitors the rolling pairwise return correlation of open positions and blocks new orders when the portfolio correlation exceeds the configured ceiling. It detects sudden hidden correlation shocks where previously independent markets begin resolving together, increasing tail risk beyond what the capital model assumes.

1.11 Risk PLANNED

CorrelationShockGuard

ModelDriftMonitor flags strategies whose live behaviour has decoupled from their backtest distribution. It computes a drift score by comparing the rolling distribution of live signals or fill prices against the expected backtest baseline and hard-rejects new orders when drift exceeds the configured ceiling, preventing a degraded model from continuing to place orders.

1.12 Risk PLANNED

ModelDriftMonitor

TailLossSimulator applies scripted adverse shock scenarios to the combined portfolio (open positions plus the proposed order) and rejects or downsizes the order if the simulated tail loss exceeds the configured maximum. It acts as a pre-trade stress test, replacing post-hoc margin calls with a forward-looking gate.

1.13 Risk PLANNED

TailLossSimulator

FeeAndGasGuard estimates the total transaction cost (CLOB maker/taker fee plus Polygon gas) for a proposed order and rejects or downsizes it when the cost-to-edge ratio exceeds the configured ceiling. It prevents strategies from placing orders where fee drag exceeds the expected edge, turning a positive-expected-value intent into a negative-EV execution.

1.14 Risk PLANNED

FeeAndGasGuard

1.15 Risk PLANNED v3.5 · wired

SettlementExposureGuard

SettlementExposureGuard tracks how much pUSD is committed in markets that share the same UMA resolution window and blocks or downsizes new orders that would push the concurrent settlement exposure above the configured ceiling. It prevents the user from having more equity at risk in a single 2-hour UMA settlement window than they can afford to lose if all markets in that window resolve adversely.

ManualOverrideAuditor intercepts every request to bypass or adjust a guardrail, enforces a rate limit on overrides per time window, requires a non-empty justification string, and emits an immutable RiskVote audit record for every approved or rejected override attempt. It ensures that manual guardrail bypasses cannot occur silently and that every override is visible to the risk team.

1.16 Risk PLANNED

ManualOverrideAuditor

1.17 Risk PLANNED v3.5 · wired

MarketHaltDetector

Watches for market-level halt conditions across Polymarket — wide-spread blowouts, missing best bid/ask, locked or crossed books, and sudden order-rate collapse. When any halt condition fires, MarketHaltDetector quarantines the affected market_id so no new OrderIntent for that market can pass the Risk pipeline. It does not pause the whole system — only the affected market. Cleared automatically once conditions normalise for a configurable cool-off window, or manually via the Admin UI.

1.18 Risk PLANNED v3.5 · wired

StaleBookGuard

Rejects any OrderIntent priced against an order book older than the configured staleness threshold. The book may look healthy, but if its last update is too old, prices have almost certainly moved. StaleBookGuard fails closed: if it cannot prove the book is fresh, it rejects.

Prevents Polytraders from trading against itself. If an outgoing OrderIntent would cross with one of our own resting orders on the same market and outcome, SelfTradeWashGuard rejects it (full overlap) or downsizes it (partial overlap). This protects against wash-trade exposure and inadvertent self-fills.

1.19 Risk PLANNED

SelfTradeWashGuard

2.1 Execution LIVE v3.5 · wired

Execution

15 modules · 4 live · 1 beta · 10 planned

Convert approved intent into executable orders. Execution never picks direction or size — those come from Strategy. It decides order type (LIMIT, IOC, FOK, POST_ONLY), timing, and microstructure behaviour (aggressive vs passive, replenish vs once-and-done). Most production bugs live here, because this is where strategy intent collides with real-world latency, fills, and re-quotes.

SmartRouter reference

SmartRouter translates an approved strategy intent into a concrete executable order by selecting the appropriate order type (FOK, GTC, or GTD), price, and timing. It may split a large order into iceberg child orders to reduce market impact. SmartRouter cannot change the direction of the trade, the target market, or the strategy intent — it is not permitted to flip the side (buy/sell), alter the outcome leg, or override any constraint set by the Risk guardrail pipeline. The only transformations it may make are to price, size scheduling, order type, and submission timing.

AntiToxicFill detects toxic order-flow patterns on the target market immediately before order submission and reshapes or cancels the ExecutionPlan to avoid being adversely selected. It monitors three primary toxicity signals: one-sided sweeps through more than three book levels in the last 5 seconds, a cancel-storm on the opposite side, and realised post-fill drift over a rolling window. If an adverse news event is detected within ±30 seconds of a planned fill, the bot enters a cooldown. Depending on signal severity, AntiToxicFill either widens the limit price by requote_widen_bps, reduces the order size by downsize_factor, or cancels the order outright. It cannot alter the trade direction, market, or outcome leg.

2.2 Execution BETA

AntiToxicFill

Execution Utility Limited live

2.3 Execution LIVE

QueueWarden

QueueWarden polices the rate at which orders and cancel-replace operations are sent to the Polymarket CLOB V2 and manages the maker queue position of resting orders. It enforces a cap on cancel-replace operations per minute to avoid CLOB rate-limit rejections, monitors drift between the resting quote price and the best bid/ask, and cancels stale resting orders that have not been filled within the stale_ttl_s window. QueueWarden also injects builderCode (bytes32) into cancel-replace messages to maintain builder attribution continuity across order lifecycle events.

NonceShepherd manages the on-chain nonce sequence for the Polytraders signing wallet on Polygon, ensuring that every V2 EIP-712 signed order carries a valid, non-colliding nonce. It detects nonce gaps (caused by dropped or reverted transactions), resequences pending orders when a gap is found, and enforces a ceiling on the number of pending signed-but-unposted orders. NonceShepherd also tracks CLOB ClobAuth credential TTLs and triggers re-authentication before they expire.

2.4 Execution LIVE

NonceShepherd

GasOracle continuously monitors the Polygon network gas-price environment and advises the execution layer on whether on-chain operations (order submission, cancel-and-replace, neg-risk conversions) are economically viable at the current gas price. It defers non-urgent operations when gas is above the rolling percentile threshold, and never defers risk-critical operations such as emergency cancellations.

2.5 Execution LIVE

GasOracle

OrderLifecycleManager is the single source of truth for every order's state machine, tracking transitions from intent through terminal state. It reconciles local state against the CLOB V2 user-state endpoint, detects stuck-in-flight and orphaned orders, and emits ExecutionReports on every transition.

2.6 Execution PLANNED

OrderLifecycleManager

FillQualityAnalyzer scores every completed fill on price slippage, execution latency, and post-fill toxicity relative to the original intent. Low-scoring fills generate WARN annotations; systematic poor scores feed back to strategy tuning.

2.7 Execution PLANNED

FillQualityAnalyzer

PartialFillHandler decides what to do with the unfilled remainder after a partial fill: hold and wait, cancel the remainder, or chase the market by submitting a new order. It respects the strategy's declared partial-fill policy and current book state.

2.8 Execution PLANNED

PartialFillHandler

CancelReplaceOptimizer selects the cheapest path to modify a resting order: amend-in-place if queue priority is preserved, or cancel-then-replace if the delta exceeds the amend threshold. It enforces rate-limit budgets across cancel and amend operations.

2.9 Execution PLANNED

CancelReplaceOptimizer

LatencyProfiler continuously measures round-trip order submission latency by route and surfaces regressions. It probes each configured route at probe_interval_s and emits ObservationReports when p95 or p99 thresholds are breached.

2.10 Execution PLANNED

LatencyProfiler

ExchangeStatusMonitor treats Polymarket itself as a degradable dependency. It polls CLOB V2 endpoint health, watches for reject-rate spikes, and parses public maintenance signals. When degradation is confirmed, it emits ObservationReports that trigger pause or de-risk actions across the exec layer.

2.11 Execution PLANNED

ExchangeStatusMonitor

PriceBandValidator is a fat-finger and decimal-shift catcher. It rejects or warns on any order priced outside a sanity band relative to the current top-of-book mid, preventing obviously erroneous prices from reaching the CLOB V2 signing step.

2.12 Execution PLANNED

PriceBandValidator

DustAndRoundingCleaner prevents creation of dust positions — holdings so small that their eventual pUSD value at settlement is less than the transaction cost of acquiring or selling them. It rounds order sizes to economically viable minimums and sweeps existing dust positions on a configurable cron schedule.

2.13 Execution PLANNED

DustAndRoundingCleaner

Stamps every approved OrderIntent with an explicit fee + slippage estimate before the order shaper sees it. The estimate is a structured object (maker_fee_bps, taker_fee_bps, expected_slippage_bps, total_cost_usd) that downstream tools and UIs all read instead of guessing.

2.14 Execution PLANNED

FeeSlippageEstimator

ExecutionUtility Spec ready

2.15 Execution PLANNED

OrderAmendCancelManager

Owns every amend and cancel operation against the CLOB for orders Polytraders has placed. Strategies do not cancel directly — they emit an OrderAdjustment intent which this bot validates, sequences, and submits via the EIP-712 v2 envelope. Maintains a strict monotonic intent sequence per order so cancels can never race amends.

ExecutionUtility Spec ready

Strategy

24 modules · 4 live · 6 beta · 14 planned

The only layer that takes directional or spread risk — every other layer supports it. Each strategy declares its risk envelope to Layer 1 at startup; a mid-run parameter change requires re-declaration and re-validation. Strategies are configuration-only by contract: the user picks a template, sets parameters, the strategy runs.

3.1 Strategy LIVE v3.5 · wired

Maker-Tight

Maker-Tight is a passive market-making strategy that posts resting maker orders inside the current touch when market conditions are benign (low volatility, adequate volume, clean order flow). It quotes both YES and NO sides at a configurable edge (edge_bps) from the mid-price, sized to clip_size_usd. When spread is too tight to earn the edge, or when order-flow imbalance signals toxic flow, the bot skips the cycle and cancels any open quotes. This is a user-controlled execution tool for posting maker liquidity on predictive-event markets — not a directional betting strategy. Maker rebates (20–25% of platform fees, paid in pUSD) contribute to the overall return profile.

Maker-Wide is a passive market-making strategy designed for thin, illiquid binary markets where the natural spread is wider than Maker-Tight's target. It posts resting maker quotes at a wider edge (80–200 bps from mid), a smaller clip size, and with a longer expected holding period. The stale-book detector (time since last fill > stale_book_minutes) prevents continuous requoting on markets with no real two-way flow. Volatility of mid over the last hour determines the spread width dynamically: higher vol → wider edge. This is a user-controlled liquidity-provision tool. Maker rebates (20–25% of platform fees, paid in pUSD) contribute to the return profile. No performance claims are made.

3.2 Strategy BETA

Maker-Wide

Sum-to-One Arb detects when the combined cost of buying both YES and NO tokens on a standard binary market falls below $1.00 pUSD, net of platform fees and a configurable fee buffer. When edge (in basis points) exceeds min_edge_bps, the bot emits a pair of OrderIntents — one for YES and one for NO — sized to the smaller of the two books up to max_leg_size_usd. Both legs fill to produce a guaranteed $1.00 settlement regardless of outcome. This is a user-controlled execution tool that exploits CLOB pricing dislocations in standard 2-outcome markets. It does not touch negative-risk or multi-outcome events; those are handled by neg-risk-sum-arb.

3.3 Strategy LIVE

Sum-to-One Arb

Neg-Risk Sum Arb exploits pricing dislocations on Polymarket’s negative-risk (multi-outcome) markets, where the sum of YES token prices across N outcomes must equal $1.00 at resolution. When sum(YES asks across outcomes) < $1.00 net of fees, the bot buys the underpriced YES tokens and — where profitable — routes the position through the NegRiskAdapter on Polygon to convert NO tokens across the set into pUSD. The 'Other' outcome is always excluded from the conversion path. This is a user-controlled execution tool that targets a structural pricing constraint specific to negative-risk event design on Polymarket. It is the multi-outcome counterpart to sum-to-one-arb.

3.4 Strategy LIVE

Neg-Risk Sum Arb

Bregman-Projection Arb detects when the joint probability distribution implied by a neg-risk multi-outcome market's order book diverges from the closest valid (simplex-constrained) distribution by more than kl_divergence_threshold nats. It uses a Frank-Wolfe iterative projection to find the nearest valid distribution, measures the KL-divergence of the observed book from that projection, and emits a set of up to max_legs_per_trade OrderIntents to exploit the divergence. This is a user-controlled execution tool for multi-outcome neg-risk markets. It never trades standard 2-outcome binary markets (those are handled by Sum-to-One Arb). No performance claims are made; the tool automates the mechanical detection and sizing of Bregman-projection violations.

3.5 Strategy BETA

Bregman-Projection Arb

Cross-Market Arb detects logically constrained relationships between pairs of Polymarket binary markets — e.g. 'Candidate A wins primary' should price at or below 'Candidate A wins general', or 'Team X advances to semi-final' should price at or below 'Team X wins tournament'. When the logical constraint is violated by more than tolerance_bps (after fee buffer), the bot emits a pair of OrderIntents (one per market) to exploit the mispricing. Market pairs must pass manual review before the bot will trade them (require_manual_pair_review=true). This is a user-controlled execution tool; it does not copy trades or make performance claims.

3.6 Strategy BETA

Cross-Market Arb

Cross-Venue Arb identifies price divergences between Polymarket and external prediction markets (Kalshi, PredictIt) that share an identical resolution source, then emits an OrderIntent to trade the cheaper leg on Polymarket. Both venues must resolve via the same authority; any mismatch blocks the trade.

3.7 Strategy PLANNED

Cross-Venue Arb

News Materiality Trader positions ahead of the book's full digestion of material news events. It receives news items from the NewsIngest pipeline (Reuters, AP, Bloomberg, league wires), scores them using a NLP materiality classifier, resolves the news entity to one or more watchlisted Polymarket markets via an entity-resolution dictionary, and emits an IOC OrderIntent when materiality_score >= materiality_threshold. A cooldown period (cooldown_s) prevents re-trading the same entity-market pair within the window. Order TTL (order_ttl_s) limits exposure if the IOC is not immediately filled. This is a user-controlled execution tool for reacting to material news before book equilibration; it does not predict resolution outcomes or make performance claims.

3.8 Strategy BETA

News Materiality Trader

Sports Model computes a quantitative fair-value probability for Polymarket sports markets using an internal power-rating model fed by league APIs (NBA, NFL, EPL, ATP/WTA, MLB), lineup and injury data via SportsFeed-Adapter, and weather data for outdoor events. When the CLOB mid-price diverges from the model price by more than min_edge_bps_vs_model, the bot sizes an IOC OrderIntent using a fractional-Kelly position sizing formula (kelly_fraction) bounded by max_per_bet_usd. The bot targets the Apr 2026 $5M pUSD maker rebate pool for sports+esports markets when the edge direction permits maker posting. This is a user-controlled execution tool that automates quantitative sports market trading decisions. No performance claims are made.

3.9 Strategy BETA

Sports Model

Resolution Fair-Value detects Polymarket binary markets that are close to resolution and priced away from their fair value (e.g. trading at ¢96 when the outcome is virtually certain). It emits an IOC OrderIntent to trade toward fair value on unambiguous markets whose oracle signal is clean and within resolution window.

3.10 Strategy PLANNED

Resolution Fair-Value

Late-Resolution Spread captures the time-decay component of high-probability outcomes near market maturity. As a binary or multi-outcome market approaches its endDate (from Gamma API), a token priced at $0.95–$0.99 retains a positive expected settlement of $1.00 if the leading outcome holds. The spread between the current market price and $1.00 shrinks deterministically as resolution approaches. This bot monitors gamma.market.endDate to identify markets within max_minutes_to_resolution of close, checks that the price-to-$1.00 spread exceeds the configurable minimum, and emits a buy OrderIntent sized to max_clip_usd. It never averages down (never_average_down is locked true). This is a user-controlled execution tool exploiting time-value decay in near-resolution predictive markets; it is not a directional price forecast.

3.11 Strategy LIVE

Late-Resolution Spread

Mean-Reversion Sniper fades price momentum in markets that have overreacted to recent information. It monitors markets where the best-ask for YES tokens exceeds price_threshold (default 0.80), computes a rolling z-score of the price change over a lookback window, and checks for order-book signals of a momentum reversal (aggressor side flips and large cancels on the heavy side). When z_score >= z_score_min and no active news cycle is detected by the NewsIngest event-density feed, the bot emits an IOC OrderIntent to sell YES tokens at or above price_threshold. An automatic time exit (time_exit_s) and stop-loss (stop_bps) control downside risk. This is a user-controlled execution tool; it does not predict outcomes or make performance claims.

3.12 Strategy BETA

Mean-Reversion Sniper

Breakout-Follower is a momentum strategy that enters Polymarket binary markets when a high-confidence news event causes a price breakout beyond configurable Bollinger Band thresholds. It scales in across multiple steps and trails a stop using an ATR multiple, acting as the directional complement to Mean-Reversion Sniper.

3.13 Strategy PLANNED

Breakout-Follower

Liquidity-Pulse is a market-analytics tool that monitors emerging Polymarket markets for early liquidity opportunities. When a market's volume and attention signal crosses configured thresholds, the bot emits an OrderIntent for user-controlled maker quote placement. All execution decisions remain with the user; this bot provides structured analytics and sizing recommendations only.

3.14 Strategy PLANNED

Liquidity-Pulse

VolatilityHarvest quotes inside the spread on volatile Polymarket event markets with thick books, collecting maker rebates while managing inventory skew. It enters when realised volatility exceeds min_realised_vol, quotes inside the current spread by quote_inside_bps, and limits total position via max_inventory_skew. After a losing quote it cools off for cool_off_after_loss seconds.

3.15 Strategy PLANNED

VolatilityHarvest

CalendarCompression identifies pairs of related Polymarket markets resolving on different dates (e.g. 'Will X happen by June?' vs 'Will X happen by December?') and trades the price differential when it deviates from the fair time-decay spread. The bot buys the underpriced near-expiry leg and/or sells the overpriced far-expiry leg to capture calendar compression.

3.16 Strategy PLANNED

CalendarCompression

ResolutionCurveTrader models the expected time-path of a Polymarket binary market's probability from the current moment to resolution, then trades when the market price deviates significantly from the model's projected curve. It captures mispricing in the temporal dimension rather than just the endpoint.

3.17 Strategy PLANNED

ResolutionCurveTrader

Dispute-Risk Pricing identifies Polymarket markets affected by UMA dispute uncertainty where the current price does not reflect the true resolution probability given the dispute risk. The bot buys positions that appear mispriced due to dispute-inflated uncertainty, but only when the resolution-rule reading is unambiguous and the parser confidence is at maximum.

3.18 Strategy PLANNED

Dispute-Risk Pricing

Rule-Risk Discount fades Polymarket binary markets where the resolution rule is ambiguous and the current market price does not reflect that ambiguity risk. When a market's price implies high certainty but the resolution rule has a significant ambiguity score, the bot takes the opposite position to capture the mean-reversion when uncertainty is priced in.

3.19 Strategy PLANNED

Rule-Risk Discount

NarrativeCrowdingFade detects Polymarket binary markets where social attention and narrative momentum have pushed prices above the statistical base rate. When attention Z-score and price drift signal crowding, the bot fades the move — taking the opposite position to capture reversion toward the base rate.

3.20 Strategy PLANNED

NarrativeCrowdingFade

FundingRotationBot manages idle pUSD capital by rotating between pre-approved safe-yield routes and Polymarket deployment, ensuring the user's deployable budget is efficiently allocated without exceeding configured reserve requirements. All rotation decisions require the user's opt-in and stay within a whitelisted route set.

3.21 Strategy PLANNED

FundingRotationBot

PortfolioHedger monitors the user's Polymarket portfolio for correlated exposure flagged by the Risk guardrail and builds offsetting positions to reduce net directional risk. When correlated market clusters exceed target_residual_corr and hedging cost is within max_hedge_cost_bps, the bot emits hedge OrderIntents to neutralise the correlation.

3.22 Strategy PLANNED

PortfolioHedger

BasisTrader identifies and trades the price basis between a Polymarket market and an off-platform reference (sportsbook line, futures price, or consensus poll) when the resolution rules of both platforms are provably identical. The bot buys the Polymarket leg when it is cheap relative to the reference and the basis exceeds min_basis_bps after fees.

3.23 Strategy PLANNED

BasisTrader

MarketCreationScout monitors newly created Polymarket markets and identifies those worth seeding with early maker liquidity before the spread tightens. It evaluates quality score, market age, and event category, then emits GTC post-only OrderIntents to seed the book with initial maker quotes, generating rebate income from early-adopter taker flow.

3.24 Strategy PLANNED

MarketCreationScout

NewsIngest continuously pulls news from external feeds (RSS, partner APIs), resolves named entities to Polymarket market IDs via the Gamma API, and scores each story for materiality. High-materiality stories trigger immediate ObservationReport emission; routine items are sampled 1/10. NewsIngest is strictly read-only — it never submits, signs, or modifies orders. Output is consumed by news-materiality-trader and contradictiondetector.

Intelligence

17 modules · 3 live · 3 beta · 11 planned

Read-only signal services. These bots never submit orders — they emit typed signals onto the internal bus that Strategy subscribes to. Every signal carries a freshness timestamp and a provenance string (feed, version, sample window) so a downstream strategy can decide whether to trust it. Stale signals must be discarded, not patched.

4.1 Intelligence LIVE

NewsIngest

OracleWatcher streams the UMA Optimistic Oracle on-chain, detecting assertion proposals, dispute filings, DVM debate escalations, and final vote outcomes for every Polymarket condition ID. It emits an ObservationReport on every state change. OracleWatcher is strictly read-only — it never submits or signs anything. Output is the primary feed for oracleriskmonitor, which uses it to gate positions during contested resolutions.

4.2 Intelligence LIVE

OracleWatcher

OrderFlowAnalyzer subscribes to ws_market for real-time trade prints and clob_public for order-book snapshots, then classifies each flow event as toxic (informed, likely to precede adverse price movement), benign (noise/retail), or informed (structural, correlated with neg-risk rebalancing). It emits an ObservationReport for every toxic flip and samples 1/10 routine ticks. Output feeds liquidity-aware strategies and the LiquidityGuard risk bot. OrderFlowAnalyzer is strictly read-only — it never submits or signs orders.

4.3 Intelligence LIVE

OrderFlowAnalyzer

CrossMarketGraph builds and maintains a directed graph of semantically equivalent and logically linked Polymarket markets by embedding market titles and rules text, applying cosine-similarity clustering, and supplementing with manual override pairs. Each node is a condition_id; each edge carries a relation_type (SAME_EVENT, COMPLEMENTARY, NEG_RISK_SIBLING, SUPERSEDES) and a confidence score. The graph is the foundation for cross-market correlation signals used by SumToOneArb and liquidity-aware strategies to detect near-duplicate markets and hedge opportunities. CrossMarketGraph is strictly read-only — it never submits or signs orders.

4.4 Intelligence BETA

CrossMarketGraph

Signal Service Limited live

4.5 Intelligence BETA

SportsFeed-Adapter

SportsFeed-Adapter ingests structured sports data from league APIs (NBA, NFL, EPL, ATP/WTA, MLB) and odds-feed providers, normalises it into a canonical SportsFeedEvent schema, and emits an ObservationReport for each qualifying update (lineup changes, injury reports, score updates, and pre-game odds shifts). It supplements primary API feeds with ws_sports for low-latency in-play state and falls back to web extraction for sports without direct API coverage. Output feeds sports-model strategies with the event data they need to price Polymarket sports markets. SportsFeed-Adapter is strictly read-only — it never submits or signs orders.

Signal Service Limited live

4.6 Intelligence PLANNED

SocialSentiment

Lightweight, secondary social-sentiment input — never primary trigger.

OnChainWatcher monitors Polygon on-chain events for a configured list of high-signal wallet addresses, detecting CTFExchangeV2 order fills, position transfers, and pUSD balance changes that exceed min_position_change_usd. It emits an ObservationReport for each qualifying wallet event after a publish_delay_s hold (to prevent front-running its own signals). Output feeds liquidity-aware strategies and the LiquidityGuard risk bot with smart-money flow intelligence. OnChainWatcher is strictly read-only — it never submits or signs orders.

4.7 Intelligence BETA

OnChainWatcher

Signal Service Limited live

4.8 Intelligence PLANNED

ResolutionRuleParser

Convert each market's resolution rule into a structured representation: source-of-truth, condition, ambiguity score.

Sanity-check that the source listed in a market's rule actually publishes what's needed for resolution.

4.9 Intelligence PLANNED

SourceOfTruthVerifier

4.10 Intelligence PLANNED

ContradictionDetector

Surface markets whose resolution rule contradicts itself, the title, or a parent neg-risk constraint.

4.11 Intelligence PLANNED

MarketOntologyBuilder

Cluster markets into events, themes, and underlying entities for cross-market reasoning.

4.12 Intelligence PLANNED

RuleChangeMonitor

Detect and emit any post-listing edit to a market's resolution rule.

4.13 Intelligence PLANNED

LiquidityForecastModel

Predict near-term book depth and trade pacing per market for sizing decisions.

4.14 Intelligence PLANNED

Wallet Flow Classifier

Score on-chain wallets on historical edge over a long window — emitted as a feature, never as a copy-trade signal.

4.15 Intelligence PLANNED

AnomalyDetector

Flag statistical anomalies in book, trades, or feed data for human review.

4.16 Intelligence PLANNED

LiquidityDecayMonitor

Tracks how quickly inside liquidity decays after our orders rest on the book. The output is a per-market liquidity-decay signal (in USD/second) that the smart router and order shaper consume to size and slice future orders. Pure observer — never blocks anything.

Signal Spec ready

4.17 Intelligence PLANNED

MarketResolutionWatcher

Watches the on-chain resolution state of every active Polymarket market. As a market approaches resolution, it emits typed warnings (T-24h, T-1h, finalising, resolved) so downstream bots can stop opening new positions and start unwinding. Pure observer — never blocks; the actual blocking is done by Risk's PreResolutionFreeze guardrail downstream.

Signal Spec ready

Security

9 modules · 0 live · 0 beta · 9 planned

Eight modules between non-custodial as a marketing word and as a real property. Non-custodial means signing happens in the user’s wallet — but every signing surface is a failure surface. These bots make that real: contract allow-listing, session scoping, signing-authority enforcement, revocation discipline, allow-lists and deny-lists, and explicit user warnings when something looks wrong.

5.1 Security PLANNED

WalletPermissionGuard

Enforce that each strategy can only call the wallet methods the user has explicitly granted, scoped per session.

Track ERC-20 allowances per token and contract; alert and shrink to a tight ceiling on idle.

5.2 Security PLANNED

AllowanceMonitor

Render a plain-English summary of every EIP-712 signature before the wallet shows the modal.

5.3 Security PLANNED

SignaturePreviewer

Issue, scope, and expire short-lived session keys so strategies can sign without re-prompting on every order.

5.4 Security PLANNED

SessionKeyManager

Nag the user to rotate signing keys on a schedule; prevent key reuse across environments.

5.5 Security PLANNED

KeyRotationReminder

Probe RPC providers continuously and fail over before a stale endpoint poisons our chain view.

5.6 Security PLANNED

RPCFailoverManager

Cross-check every order’s chain-derived inputs (nonce, balance, allowance) against multiple sources before signing.

5.7 Security PLANNED

ChainStateVerifier

ContractAddressGuard reference

5.8 Security PLANNED

ContractAddressGuard refuses to allow a signature or on-chain submission against any contract address that is not present on the committed CLOB V2 allow-list. It enforces the EIP-712 domain separator match against the expected V2 domain and rejects any order that targets a V1 Exchange address. This is a hard security control for the V1-to-V2 migration and must not be loosened without an explicit signed-off admin change. It cannot modify orders — it only approves or rejects.

Guardrail Ready to build

5.9 Security PLANNED

WalletFundingGuard

Rejects any OrderIntent whose required pUSD collateral cannot be covered by the funded balance of the assigned wallet, including a configurable buffer. Prevents the system from submitting orders that would fail at the exchange for insufficient funds, which burns latency and creates noisy reject metrics.