No user-submitted related servers found.
Scored across 34 tools
Most tools target clearly distinct operations, and the very detailed descriptions carefully separate recall/reconstruct/recall_with_context and check_health/get_session_findings/deep_check. However, the recall variants and the two overlapping health-reporting seams mean an agent must read carefully to choose correctly.
All tool names use consistent snake_case, overwhelmingly in a verb_noun pattern (store, recall, get_profile, delete_memory, pause_persistence, set_recall_precision). There is no camelCase or mixed-convention drift, and the few bare verbs (recall, store) remain readable and conventional.
34 tools is well above the 3–15 sweet spot and past the 25-tool threshold, making the surface heavy for an agent to navigate even if most tools have distinct purposes. Several capabilities are split into read/write/status trios (pause/resume/persistence_status, set/get_recall_precision), which adds further bulk.
The surface covers memory CRUD, episodes, profiles, declared associations, recall/reconstruction, health checks, persistence control, precision tuning, import/export/merge, migration, and update checks — a broad lifecycle. Minor gaps remain, such as no standalone delete_profile, but core workflows are well covered.
