Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx skills add hmesutozsoy/oddsrail --skill check-cross-venue-edgegit clone --depth 1 https://github.com/hmesutozsoy/oddsrailWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/hmesutozsoy/oddsrail/check-cross-venue-edge)<a href="https://agentmods.dev/skills/hmesutozsoy/oddsrail/check-cross-venue-edge"><img src="https://agentmods.dev/badge/skills/hmesutozsoy/oddsrail/check-cross-venue-edge/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/hmesutozsoy/oddsrail/check-cross-venue-edge"><img src="https://agentmods.dev/badge/skills/hmesutozsoy/oddsrail/check-cross-venue-edge.svg" alt="Reviewed on agentmods" width="80" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00044 | $0.00330 |
| Opus 5.5 | $0.00018 | $0.00132 |
| Sonnet 5.5 | $0.00009 | $0.00066 |
| Haiku 4.5 | $0.00004 | $0.00033 |
Grade A, and why
check-cross-venue-edge scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured today.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
What it actually says
check cross venue edge
Use the tools of the oddsrail MCP server (install: pip install oddsrail, then configure your MCP client with local stdio command oddsrail and no arguments). Prices are implied probabilities in (0,1). Trading tools are dry-run unless the operator has set ODDSRAIL_DRY_RUN=0; never assume an order was placed without checking the dry_run field of the result.
Investigate whether '$ARGUMENTS' offers a genuine cross-venue edge.
- compare_venues('$ARGUMENTS'): candidates only. It returns nothing for most queries, which is the honest answer, not a failure.
- For any candidate: settlement_audit(polymarket_id, kalshi_ticker). A 'block' verdict ends it: the legs do not hedge each other and the price difference is not an edge.
- On 'ok' or 'caution': quote_cost on BOTH legs at the size you would actually trade. Cross-venue differences are usually smaller than the combined slippage.
- Read resolution_criteria on both sides yourself. Identical wording does not mean identical settlement.
- Only then size the trade.
State plainly if the conclusion is 'no edge here'. That is the usual and correct outcome.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- today Changed 72d2d7753cf4
- 11d ago Changed 732d33168149
- 1mo ago First seen · 26 lines · 44 tokens per session scan A 879c1a045b45
check-cross-venue-edge is a skill published in the GitHub repository hmesutozsoy/oddsrail (1 stars, last pushed today), licensed MIT. It adds 44 tokens to every session and 330 once invoked, about $0.0002 per session on Opus 5.5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-09-04.
Other skills, from other repositories
polymarket-copytrading
Mirror Polymarket whale wallets, inspect wallet positions, follow specific trader addresses, or paper-trade copytrading with $SIM before using real money. Polling mode is free; Reactor mode is Pro for event-driven real-time mirroring via Simmer's on-chain signal infrastructure.
polymarket-weather-trader
Trade Polymarket temperature markets from NOAA and Open-Meteo forecasts via the Simmer API. Use to run or configure a weather strategy, or to check the forecast a market resolves on.
polymarket-btc-up-down-trader
Trade Polymarket BTC daily and weekly UP/DOWN markets with empirically-anchored exit discipline. Enters on CEX momentum divergence; exits automatically on time cap, volume spike, or target capture. Use when the user wants to trade BTC direction markets (hours/days duration), not fast 5-minute markets.
polymarket-fast-loop
Trade Polymarket 5-minute and 15-minute crypto sprint markets, automate short-term prediction trading, use CEX momentum or custom signals, or monitor sprint-market positions. Default signal is Binance BTC/USDT klines.
polymarket-soccer-shock-ladder
Fade sharp in-play price shocks on Polymarket soccer markets with a laddered limit-buy strategy (Roan's FIFA-quant framework). Pro skill. Currently scoped to 2026 World Cup markets. Simmer's server detects shocks in real time and emits pre-sized signals; this skill places the recovery ladder and manages the exit.
nansen-copytrader-overlay
Rank Polymarket copytrading leaders using Nansen as a data layer — pnl-by-market for the target market is the primary signal, address-summary/pnl-by-address is a secondary, credit-guarded refinement. Includes an experimental (dry-run only) live insider scan. Use when Simmer needs to re-rank or filter a leader list…