Memory MCP Server
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
No arguments | |||
Instructions
Guidance the server publishes about itself, which clients place ahead of the tool catalog so the model reads it before choosing anything.
This server publishes no instructions, or was last inspected before Glama recorded them.
Capabilities
Server capabilities have not been inspected yet.
Tools
Functions exposed to the LLM to take actions
| Name | Description |
|---|---|
| add_short_term_memoryB | Add a new short-term memory entry from recent conversation messages. The memory will be indexed by keywords and scored based on relevance over time. |
| search_short_term_memoriesB | Search and retrieve relevant short-term memories based on recent conversation context. Returns top relevant, next relevant, and random flashback memories. |
| delete_short_term_memoriesC | Delete short-term memories matching a keyword or regex pattern. |
| get_memory_statsC | Get statistical information about short-term memories. |
| cleanup_memoriesA | Manually trigger cleanup of old or low-relevance short-term memories. This removes memories older than 1 year or with very low relevance scores, keeping at least 512 memories. |
| get_frequent_conversationB | Get the most frequently mentioned conversation ID in memories. |
| add_long_term_memoryC | Add a new long-term memory with a trigger condition. The trigger is JavaScript code that determines when this memory should be activated. Available context: context.messages (array), context.conversation_id (string), context.participants (object). Available functions: match_keys(messages, keywords, scope, depth), match_keys_all(messages, keywords, scope, depth). |
| update_long_term_memoryC | Update an existing long-term memory. You can update the trigger condition, prompt content, or add update context. |
| delete_long_term_memoryC | Delete a long-term memory by name. |
| list_long_term_memoriesC | List all long-term memory names and their basic information. |
| search_long_term_memoriesC | Search and activate relevant long-term memories based on current conversation context. Returns activated memories (whose triggers evaluated to true) and random memories for serendipity. |
| get_memory_contextC | Get the creation and update context of a specific long-term memory. |
| backup_memoriesB | 將指定對話的所有記憶備份到文件。支持導出短期和長期記憶，包含完整的元數據和時間戳。 |
| restore_memoriesA | 從備份文件還原記憶。警告：這將覆蓋當前對話的所有記憶。 |
| list_backupsC | 列出可用的備份文件 |
| delete_backupC | 刪除指定的備份文件 |
| search_memoriesC | 使用高級過濾條件搜索記憶。支持關鍵詞、時間範圍、分數過濾等。 |
| analyze_memory_patternsC | 分析記憶使用模式，提供統計信息和洞察 |
| health_checkC | 獲取服務器健康狀態和性能指標 |
| get_metricsC | 獲取服務器性能指標 |
| get_cache_statsC | 獲取查詢緩存統計信息 |
Prompts
Interactive templates invoked by user choice
| Name | Description |
|---|---|
| remember-user-info | Store important user information in long-term memory |
| recall-context | Search for relevant memories based on current conversation |
| create-reminder | Create a conditional reminder that activates based on context or date |
| analyze-conversation | Analyze conversation history and create appropriate memories |
Resources
Contextual data attached and managed by the client
| Name | Description |
|---|---|
| Memory System Overview | Overall statistics and health of the memory system |
| Conversations List | List all conversations with stored memories |
| Conversation Statistics | Detailed statistics for a specific conversation |
| Memory Usage Best Practices | Guide on how to effectively use the memory system |
TDQS
Scored across 21 tools
Most tools have distinct purposes targeting specific memory types (long-term vs. short-term) or operations (add, delete, search, backup), but some overlap exists: 'search_memories' and 'search_short_term_memories' could be confused, and 'get_metrics' and 'health_check' both relate to performance monitoring. Descriptions help clarify, but boundaries are not perfectly clear.
The naming is mixed with no consistent pattern: some use verb_noun (e.g., 'add_long_term_memory', 'delete_short_term_memories'), others use noun_verb (e.g., 'health_check'), and there's a mix of English and Chinese names. While readable, the conventions vary significantly across the tool set.
With 21 tools, the count is slightly high but reasonable for a memory management server covering operations like CRUD, search, backup, and analytics. It feels comprehensive without being overly bloated, though it could be streamlined by merging some overlapping tools.
The tool set provides complete coverage for memory management: CRUD operations for both long-term and short-term memories, search functionalities, backup/restore, cleanup, and analytics (e.g., stats, patterns). No obvious gaps exist; agents can handle the full lifecycle of memories effectively.