auto-bean-memory

To manage reusable beancount memory records, inspect, review, and persist durable advisory context tied to import and categorization workflows. This skill stores updates with deterministic JSON formatting, preserving provenance and audit context for future ledger decisions.

Adding audit trail context to fixed memory stores requires persisting decisions with structured metadata fields like schema_version, memory_type, source, and review_state. This approach ensures governance and auditability across workspace memory records.

Pruning memory records without losing provenance involves reorganizing durable advisory context while preserving the audit fields and source history. This skill maintains structured metadata and deterministic formatting to keep memory small, reviewable, and tied to evidence.

Persisting reusable memory decisions for categorization requires satisfying schema_version, memory_type, source, decision, scope, confidence, review_state, created_at, updated_at, and audit fields. These fields ensure deterministic JSON formatting and provenance context.

Reviewing fixed memory store records is necessary when inspecting, correcting, or reorganizing durable advisory context created by import, categorization, naming conventions, and related governance actions across ledger workspaces.