chimera_score
Score messages 0–1 for context-window compression or goal alignment. Use drop_priority to evict lowest scores, or importance_for_goal to retain messages aligned with your focus.
Instructions
Score messages 0–1 for context-window management. mode='drop_priority' (default): scores by recency+type+density — lowest scores are dropped first in lossy compression. mode='importance_for_goal': scores by alignment with the focus goal — highest scores are most relevant to keep. Vs. direct reasoning: O(n) tokenisation is far cheaper than asking the model to rank N messages, which consumes O(N*content) prompt tokens.
Input Schema
TableJSON Schema
| Name | Required | Description | Default |
|---|---|---|---|
| messages | Yes | Messages to score [{role, content}] | |
| focus | No | Task focus/query. Required for importance_for_goal mode; improves drop_priority scoring. | |
| mode | No | drop_priority: score for compression — lowest scores evicted first. importance_for_goal: score by focus alignment — highest scores most relevant. | drop_priority |