Kalshi Fed Speech Signal Trader#

This is a template. The default signal uses static keyword dictionaries -- remix it with NLP sentiment models, live Fed speech transcripts via FRED API, or real-time news feeds. The skill handles all the plumbing (market discovery, trade execution, safeguards). Your agent provides the alpha.

Strategy Overview#

Fed speeches contain hawkish and dovish signals that predict rate decisions. This skill scores net sentiment from keyword matching on market question text, then adjusts the fair probability of a rate cut. When the adjustment creates a gap vs. rate cut market prices, it trades.

Key advantages:

• No external data needed -- extracts signal from market question text itself
• Extensible -- add new keywords, adjust weights, or plug in NLP models
• Cross-signal aggregation -- pools sentiment across all Fed-related markets

Signal Logic#

Sentiment Scoring#

1. Scan all Fed rate market questions for hawkish/dovish keywords
2. Weight matches (some keywords stronger signals than others)
3. Compute net sentiment: dovish_total - hawkish_total
4. Adjust baseline cut probability by 5% per net unit
5. Trade rate cut markets when |fair - market| >= entry_edge

Keyword Dictionaries#

Hawkish (reduce cut probability): "inflation persistent", "tightening", "restrictive", "price stability", "higher for longer", etc.

Dovish (increase cut probability): "data dependent", "labor softening", "gradual", "balanced", "appropriate to reduce", etc.

Conviction-Based Sizing#

• conviction = min(|edge| / entry_edge, 2.0) / 2.0
• size = max($1.00, conviction * MAX_POSITION_USD)
• Larger edge = larger position, capped at MAX_POSITION_USD

Risk Parameters#

ParameterDefaultNotesEntry edge10%Min fair-vs-market divergence to tradeExit threshold45%Sell when position price reaches thisMax position size$5.00 USDCPer marketMax trades per run3Rate limitingMax slippage15%Skip if slippage exceedsMin liquidity$0Disabled by default

Installation & Setup#

bash

clawhub install kalshi-fed-speech-signal-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.

ScenarioModeFinancial riskpython trader.pyDry runNoneCron / automatonDry runNonepython trader.py --liveLive (Kalshi via DFlow)Real USDC

Required Credentials#

VariableRequiredNotesSIMMER_API_KEYYesTrading authority. Treat as a high-value credential.SOLANA_PRIVATE_KEYYesBase58-encoded Solana private key for live trading.

Tunables (Risk Parameters)#

VariableDefaultPurposeSIMMER_FED_SPEECH_ENTRY_EDGE``0.10Min divergence to trigger tradeSIMMER_FED_SPEECH_EXIT_THRESHOLD``0.45Sell position when price reaches this levelSIMMER_FED_SPEECH_MAX_POSITION_USD``5.00Max USDC per tradeSIMMER_FED_SPEECH_MAX_TRADES_PER_RUN``3Max trades per execution cycleSIMMER_FED_SPEECH_SLIPPAGE_MAX``0.15Max slippage before skipping tradeSIMMER_FED_SPEECH_MIN_LIQUIDITY``0Min 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: hello@simmer.markets

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