LLM Game Engine Bridges AI And Narrative Games
A new LLM Game Engine framework bridges traditional narrative engines (Ren'Py, Twine) and AI chat interfaces by combining a causal graph and stat system to enforce deterministic story progression. Built in Python 3.11 and using Anthropic's Claude Haiku for the classifier loop, it includes an example 'escape_room' story and JSON-based story formats. The design preserves free-form player input while only matched actions and bounded ±5 stat changes advance the narrative.
Key Points
- 1Implements a causal graph of story nodes to enforce deterministic, authorial-locked progression
- 2Limits LLM influence by validating stat changes and clamping ±5 per interaction for safety
- 3Enables player free-form input while ensuring only matched actions advance narrative and stats
Scoring Rationale
Actionable, open-source engine enforcing deterministic progression; moderate novelty combining known causal-graph and stat-gating patterns.
Sources
Public references used for this report.
Practice with real Logistics & Shipping data
90 SQL & Python problems · 15 industry datasets
250 free problems · No credit card
See all Logistics & Shipping problems
