review-only-evaluates-derived-beliefs
IN premise
`review_beliefs` filters out premises (nodes without justifications); only derived beliefs with at least one justification are sent for LLM review.
Summary
When you ask the system to review its beliefs for correctness, it only checks the ones it derived from other beliefs — not the foundational observations or facts that were entered directly. This means premises are treated as ground truth that the review process won't question, so if a premise is wrong, the review command won't catch it.
Dependents
These beliefs depend on this one:
- review-pipeline-is-scoped-and-mutation-safe — The belief review pipeline restricts evaluation to derived beliefs only (premises excluded) and gates auto-retraction behind the dry-run flag, ensuring review operations are scope-limited and mutation-safe by default.
Details
| Source | entries/2026/05/05/reasons_lib-review.md |