A
licenseNot graded
qualityA
maintenancePrivacy Agent Memory with best performance and security
35
MIT
knowmind
Related Servers
Alternatives to knowmind
No user-submitted related servers found.
Related Servers
- FlicenseNot gradedqualityBmaintenanceLong-term Memory for AI. On Device. Secure. Coding tools, AI Agents. Instant Recall. Precise.-
- AlicenseAqualityDmaintenanceEnables AI agents with persistent semantic memory, including semantic recall, knowledge graphs, and instant domain expertise via pre-built Intelligence Packs.1032 npmMIT
- FlicenseNot gradedqualityBmaintenanceA long-term memory system built for AI Agents. Agent wakes up already knowing who he is, not querying "who am I?" every session. Every turn calling back accurate memory context. Achieving accurate memory hits while also preventing memory from expanding at scale. No compression, no forgetting.38-
- AlicenseAqualityAmaintenancePersistent AI agents memory for Claude, Codex, Cursor... that is always current.29249 npm102Apache 2.0
- AlicenseNot gradedqualityDmaintenanceEnables AI agents to maintain persistent, local memory with retrieval-augmented search, knowledge graphs, and context surfacing, without any cloud dependencies.439 npmMIT
TDQS
A4.2/5.0
Scored across 22 tools
Disambiguation4/5
Among the active tools, purposes are distinct: store/upload split by content length, recall vs recall_at_time split by time semantics, entity vs link vs unlink, and inferred-review pair. However, 7 tools (recall, health, stats, entity, link, schema, unlink) are deprecated aliases that still answer, creating seven near-duplicate pairs an agent could misroute to if it skims.
Naming Consistency4/5
Active tools follow a clean knowmind_verb_noun snake_case pattern (get_health, store_memory, link_entities, recall_at_time). The deprecated aliases (knowmind_health, knowmind_stats, knowmind_entity, knowmind_link) drop the verb and break the pattern, a minor but visible inconsistency.
Tool Count3/5
22 tools is on the heavy side; 7 of them are deprecated aliases, so the effectively supported surface is ~15, which is reasonable for a memory-plus-graph server. The alias inflation pushes the raw count into the borderline-heavy band.
Completeness4/5
Coverage is broad: write (store/upload), supersede (update_fact), time-travel and current recall, listing (recent/relations), entity upsert, link/unlink, inferred-relation review, plus health/stats/schema diagnostics. Intentional no-hard-delete design and upsert-by-title document replacement leave only minor gaps (no batch ops, no explicit document edit beyond re-upload).
Maintenance
ActivityActive
ResponsivenessNo issues