{"id":"context-opacity","text":"The human cannot track what the LLM currently has in context. Context windows are opaque — the human sees their messages and tool results but not the model's internal state, compaction decisions, or what was silently dropped.","truth_value":"IN","source":"repo:beliefs-pi/CLAUDE.md","source_url":"","source_hash":"","justifications":[],"dependents":["continuity-human-problem"],"metadata":{},"created_at":"","updated_at":"","reviewed_at":"","verified_at":"","retracted_at":"","explanation":{"steps":[{"node":"context-opacity","truth_value":"IN","reason":"premise"}]}}