compound-memory
Related Servers
Alternatives to compound-memory
No user-submitted related servers found.
Related Servers
- AlicenseNot gradedqualityCmaintenanceGives AI agents a durable, locally hosted temporal memory built from a markdown vault you own, exposing hybrid semantic/lexical/entity search over events, threads, transcripts, entity relationships with validity dates, and patterns with confidence decay. Writes are approval-gated, so agents can draft, commit, and sync new memories while keeping every index rebuildable and nothing leaving your machine.MIT
- FlicenseNot gradedqualityCmaintenanceProvides AI agents with persistent, local cross-session shared memory by combining vector semantic retrieval with knowledge graph relationships, and supports short/long-term memory management and local backups.-
- AlicenseAqualityBmaintenanceEnables AI coding agents to share a persistent, local-first memory of architectural decisions, conventions, bug root causes, and session handoffs, searchable via full-text ranking, so context carries across tools and sessions without any cloud dependency.213437 npm2MIT
- AlicenseNot gradedqualityCmaintenanceEnables AI agents to store and recall durable facts, decisions, and lessons over MCP using hybrid semantic and keyword search, typed memories, graph relations, sessions, and time-travel queries. Runs locally with no server process or API key, persisting everything to files that both agents and a CLI can share.28 npmMIT
- AlicenseAqualityCmaintenanceEnables coding agents to share a persistent local memory store across sessions and clients, saving, searching, updating, and retrieving facts, decisions, incidents, preferences, references, and snippets with semantic and lexical search.10MIT
- AlicenseNot gradedqualityBmaintenanceGives every MCP-capable AI agent a shared, local-first memory store, letting them save, search, update, and delete preferences, decisions, and facts kept as plain Markdown files versioned in git. Memories are ranked by relevance on recall and shared across Claude Code, Cursor, Codex, and any other MCP client.MIT
TDQS
Scored across 5 tools
Each tool maps to a distinct memory operation: search (query-based recall), get (id-based fetch), write, feedback, and link. The search-vs-get boundary is clear because one takes a query and the other takes an id, and their descriptions reinforce the distinction. No two tools appear to do the same thing.
All tools follow a strict `memory_<verb>` snake_case pattern: memory_search, memory_write, memory_feedback, memory_get, memory_link. The shared prefix plus a distinct action noun makes the whole set predictable and readable at a glance.
Five tools is well-scoped for a memory store: write, read by id, search, link, and feedback cover the intended loop without padding. No tool feels redundant or out of place, and nothing suggests a missing primitive that would require an extra tool.
The surface covers create, read, search, associate, and outcome-based lifecycle (obsolete archives), which is strong for the domain. Gaps are minor: there is no explicit delete/forget or edit tool, so removal relies on `memory_feedback(outcome='obsolete')`, and conflict resolution via keys is implied rather than directly operable.