Kalshi Econ Bin Sum Trader

Trades CPI range bin markets on Kalshi using the constraint that mutually exclusive bins must sum to ~100%. Normalizes and trades the most mispriced bin when...

Security Scan

The skill largely matches a Kalshi trading template but contains internal inconsistencies about required credentials and a truncated code file — it asks for highly sensitive keys (SIMMER_API_KEY and a Solana private key) that are plausible for live trading but need careful review before use.

Purpose & Capability

The skill's name and description claim a Kalshi/CPI bin-sum trading strategy and the code and docs call out using simmer-sdk to interact with Kalshi — that is coherent. However the registry metadata presented to the evaluator initially listed no required env vars while SKILL.md and clawhub.json explicitly require SIMMER_API_KEY and SOLANA_PRIVATE_KEY. That metadata mismatch is an inconsistency that should be resolved.

Instruction Scope

SKILL.md and trader.py instruct the agent to use the SIMMER_API_KEY and (in some sections) a SOLANA_PRIVATE_KEY for live trading. The instructions reference importing markets, executing trades, and integrating with DFlow/Solana — all networked actions. There is no instruction to read unrelated local files, but the provided trader.py is truncated in the package listing ("…[truncated]") so parts of runtime behavior are not visible; that missing code could change the assessment.

Install Mechanism

This is instruction-only from the platform perspective (no custom binary install). Dependencies are limited to simmer-sdk (PyPI) and optional tradejournal; requiring a PyPI package is expected for an SDK-based trader. There is no external arbitrary-download install step in the manifest.

Credentials

The skill requires SIMMER_API_KEY (expected for the Simmer SDK) and requests a SOLANA_PRIVATE_KEY for live trading; a Solana private key grants broad signing power and is highly sensitive. These credentials are plausible for executing live trades via a Solana-based settlement flow, but the README/metadata inconsistency and lack of an explicit explanation of why Kalshi trades require a Solana key increases risk. Ensure the Solana key scope and how it is used are audited before supplying it.

Persistence & Privilege

The skill does not request always:true, autostart is false, and automaton entrypoint is present but not started by default. Autonomous invocation is allowed (platform default), which is normal for skills; there is no evidence of the skill modifying other skills or global agent settings.

What to consider before installing

This skill appears to be a reasonable trading template, but there are red flags you should address before installing or giving it real credentials: 1) Metadata mismatch — the registry summary showed no required envs, but SKILL.md and clawhub.json require SIMMER_API_KEY and SOLANA_PRIVATE_KEY; confirm which is authoritative. 2) The SOLANA_PRIVATE_KEY is highly sensitive — only provide it if you fully trust the simmer-sdk implementation and understand how the key is used and stored. 3) Review the simmer-sdk PyPI project and its GitHub source yourself (or have a developer audit it) to ensure the SDK doesn't forward keys to unexpected endpoints. 4) The included trader.py in the listing was truncated; request the full source and inspect the remaining code paths for any data exfiltration, unexpected network calls, or opaque subprocess execution before running live. 5) Test in dry-run mode only, with minimal funds, and prefer ephemeral or restricted keys if possible (e.g., an API key with only test/paper permissions). If you can't audit the remaining code, treat the skill as untrusted for live trading.

This is a template.
The default signal uses the fundamental constraint that mutually exclusive outcome bins must sum to 100% -- remix it with Bayesian priors, consensus forecasts, or cross-event correlation models.
The skill handles all the plumbing (market discovery, trade execution, safeguards). Your agent provides the alpha.

Strategy Overview

CPI range bin markets on Kalshi price each outcome bin independently. Since exactly one bin must resolve YES, the probabilities must sum to 100%. When the market sum deviates beyond tolerance, at least one bin is mispriced. This skill normalizes the distribution and trades the most mispriced bin.

Key advantages:

• Mathematical certainty -- bins MUST sum to 100%, any deviation is a guaranteed mispricing
• No forecasting needed -- the strategy is model-free, relying purely on arbitrage math
• Self-correcting -- as the sum approaches 100%, signals disappear (no stale trades)
• Works across any bin-style market -- CPI, GDP, unemployment, any mutually exclusive set

Signal Logic

Bin Sum Normalization

1. Group CPI bin markets by event (e.g., "March 2026 CPI")
2. Sum all bin prices in the group
3. If |sum - 1.0| > sum_tolerance, normalize: fair_prob = price / sum
4. Compute edge per bin: edge = fair_prob - market_price
5. Trade the bin with the largest absolute edge

Example

Bin	Market Price	Fair (normalized)	Edge	Action
CPI < 2.0%	5%	4.8%	-0.2%	Hold
CPI 2.0-2.5%	25%	23.8%	-1.2%	Hold
CPI 2.5-3.0%	35%	33.3%	-1.7%	BUY NO
CPI 3.0-3.5%	30%	28.6%	-1.4%	Hold
CPI > 3.5%	10%	9.5%	-0.5%	Hold
Sum	105%	100%

Conviction-Based Sizing

• conviction = min(|edge| / deviation, 2.0) / 2.0
• size = max($1.00, conviction * MAX_POSITION_USD)
• Larger edge relative to total deviation = larger position

Remix Ideas

• Cross-event sum: Apply same logic to GDP, unemployment, or Fed rate bins
• Consensus prior: Weight normalization toward consensus CPI forecasts
• Multi-bin arbitrage: Trade multiple bins simultaneously for hedged positions
• Rolling rebalance: Re-run as market moves to capture dynamic mispricing

Risk Parameters

Parameter	Default	Notes
Sum tolerance	5%	Min deviation from 100% before trading
Exit threshold	45%	Sell when position price reaches this
Max position size	$5.00 USDC	Per market
Max trades per run	3	Rate limiting
Max slippage	15%	Skip if slippage exceeds
Min liquidity	$0	Disabled by default

Installation & Setup

clawhub install kalshi-econ-bin-sum-trader

Requires: SIMMER_API_KEY and SOLANA_PRIVATE_KEY environment variables.

Cron Schedule

Cron is set to null -- the skill does not run on a schedule until you configure it in the Simmer UI.

Safety & Execution Mode

The skill defaults to dry-run mode. Real trades only execute when --live is passed explicitly.

Scenario	Mode	Financial risk
python trader.py	Dry run	None
Cron / automaton	Dry run	None
python trader.py --live	Live (Kalshi via DFlow)	Real USDC

The automaton cron is set to null -- it does not run on a schedule until you configure it in the Simmer UI. autostart: false means it won't start automatically on install.

Required Credentials

Variable	Required	Notes
SIMMER_API_KEY	Yes	Trading authority. Treat as a high-value credential.
SOLANA_PRIVATE_KEY	Yes	Base58-encoded Solana private key for live trading.

Tunables (Risk Parameters)

All risk parameters are declared in clawhub.json as tunables and adjustable from the Simmer UI without code changes.

Variable	Default	Purpose
SIMMER_ECON_BINSUM_SUM_TOLERANCE	0.05	Min deviation from 100% sum before trading
SIMMER_ECON_BINSUM_EXIT_THRESHOLD	0.45	Sell position when price reaches this level
SIMMER_ECON_BINSUM_MAX_POSITION_USD	5.00	Max USDC per trade
SIMMER_ECON_BINSUM_MAX_TRADES_PER_RUN	3	Max trades per execution cycle
SIMMER_ECON_BINSUM_SLIPPAGE_MAX	0.15	Max slippage before skipping (0.15 = 15%)
SIMMER_ECON_BINSUM_MIN_LIQUIDITY	0	Min market liquidity USD (0 = disabled)

Dependency

simmer-sdk is published on PyPI by Simmer Markets.

• PyPI: https://pypi.org/project/simmer-sdk/
• GitHub: https://github.com/SpartanLabsXyz/simmer-sdk
• Publisher: [e-mail]

Review the source before providing live credentials if you require full auditability.