CogniCore
v0.7
GridWorld-v1
CodeDebug-v1
MultiAgent-v1
NPCSim-v1
Workflow-v1
8×8
10×10
15×15
1×
3×
10×
Agent
Heatmap
Policy
Heat+Policy
Memory
Reflect
▶ Train
↻
Ready
📈
Training Progress
🧠
Agent Memory
💡
Reflection Engine
0%
Win Rate
0
Avg Reward
—
Best
0
Q-Table
0
Mem Hits
0
Reflections
—
Mistakes ↓
⚡
Learning Phase
Exploring
Learning
Adapting
Mastered
0
Episode
0
Steps
0
Reward
—
Best
1.00
Epsilon
0
Q-States
0%
Win Rate
Ready
Phase