Ignition is a study-mode poker assistant focused on Ignition Casino style NLHE cash games. It provides a lightweight command line coach for preflop range work and simple post-flop heuristics so that players can review hands away from the tables and stay compliant with site terms of service.
- Preflop range lookup for 6-max 100bb games with curated open, defend and re-raise charts.
- Post-flop heuristic engine that classifies hand strength, recognises basic draws and suggests actions using a discrete action set.
- Estimated equity versus a solid opponent range and short rationale text for each recommendation to encourage understanding instead of pure memorisation.
The assistant is implemented in pure Python. Run the CLI from the repository root:
python -m ignition.cli --stage preflop --position CO --stack 100 --hand AsKd
Example post-flop usage:
python -m ignition.cli \
--stage flop \
--position BTN \
--stack 100 \
--pot 9 \
--hand AsQs \
--board Td8s3s
Flag | Description |
---|---|
--stage |
Game stage (preflop , flop , turn , river ). |
--position |
Hero position (UTG , HJ , CO , BTN , SB , BB ). |
--stack |
Effective stack in big blinds (used for SPR and range selection). |
--pot |
Pot size in big blinds (post-flop only). |
--hand |
Hero hole cards, e.g. AsKd . |
--board |
Community cards, e.g. AhKdTs for the flop. |
--situation |
Preflop context (open , vs_raise , vs_3bet ). |
ignition/cards.py
– parsing helpers and canonical hand notation.ignition/range_engine.py
– loads curated JSON ranges and returns preflop actions.ignition/postflop.py
– board texture analysis and heuristic recommendations.ignition/poker_assistant.py
– high-level facade combining both modules.ignition/data/preflop_ranges.json
– 100bb 6-max preflop ranges used by the coach.
The current version is designed for offline study. Future improvements could include:
- Additional stack depths (40bb, 60bb) and tournament antes.
- Turn and river node solving distilled from full solvers to refine the heuristic engine.
- Range visualisation UI and heatmaps backed by the JSON data.
- Opt-in hand history logging for personalised leak analysis.