servers / aimemory-mcp-server
aimemory-mcp-server
communityunknownpythondestructive capablehealthy
AI Memory MCP Server — persistent memory for AI assistants via Model Context Protocol. Cross-platform search across ChatGPT, Claude, DeepSeek, Gemini & Kimi.
01Tools · 13
How to read this: tool names here are observed from a live tools/list handshake. The Risk label is a heuristic inferred from the tool name (write/destructive verbs), not from executing the tool — a conservative guess, not a verified capability. We never escalate risk from a description. Found one that's wrong? Tell us — we fix on report.
| Tool | Risk | Side effects | Approval |
|---|---|---|---|
| import_memories Import memories from a JSON list. Useful for restoring backup or bulk import. | read | false | unknown |
| save_memory Save a new memory to your knowledge base.
Use this to store important facts, preferences, code snippets, meeting notes,
or any conversation takeaways that you want to remember for future sessions.
The memory will be searchable via full-text search (FTS5) and can be
retrieved by any MCP-connected AI assistant.
Examples of what to save:
- "User prefers Python over JavaScript for backend work" (preference)
- "Team uses PostgreSQL for all new projects" (fact)
- "```python\ndef hello(): print('hi')```" (code snippet) | read | false | unknown |
| list_memories List memories, optionally filtered by tag. Returns newest first. | read | false | unknown |
| update_memory Update an existing memory's content and/or tags. | write | true | unknown |
| get_memory Retrieve a single memory by its ID. | read | false | unknown |
| search_memories Search your memory store using full-text search (FTS5).
Searches across content, tags, and source fields. Results are ranked by
relevance score. This is the primary way your AI assistant retrieves
past information — use it to find relevant context from conversations
across ChatGPT, Claude, DeepSeek, Gemini, Kimi, and more. | read | false | unknown |
| clear_all_memories Delete ALL memories from the database. This action cannot be undone!
⚠️ WARNING: This permanently removes every memory. Export your data first
using export_memories if you want to keep a backup.
Returns:
A dict with "success" boolean and "deleted_count" (number of memories removed). | read | false | unknown |
| memory_stats Get statistics about your memory store.
Returns:
A dict with total count, tag distribution, and recent activity summary. | read | false | unknown |
| get_all_tags Get all unique tags from your memory store with usage counts.
Returns:
A dict with "total_tags" count and "tags" mapping of tag name to usage count,
sorted by most-used first. Example: {"total_tags": 12, "tags": {"work": 45, "project": 32, ...}}
Use this to discover what categories of memories you have stored. Great for
building navigation, understanding your knowledge base structure, or finding
underused tags. | read | false | unknown |
| inject_memory Search relevant memories and return them formatted for AI context injection (CORE FEATURE).
This is the core "memory injection" feature — it finds memories relevant to the
current conversation and formats them so they can be injected into the AI's context.
Use this at the start of conversations or when you need relevant background.
WHY THIS MATTERS: Most MCP memory servers just store/search. AI Memory formats
the results READY for injection — saving you prompt engineering work. | read | false | unknown |
| export_memories Export all memories as JSON for backup or migration.
Returns:
A dict with "count" (number of memories) and "memories" (list of all memory objects).
Use this to backup your knowledge base or migrate to another system. | read | false | unknown |
| delete_memory Delete a memory by its ID. | destructive | true | true |
| batch_save_memories Save multiple new memories at once. Optimized for bulk operations and AI conversation analysis. | unknown | unknown | unknown |
02Install & source
uvx aimemory-mcp-server
uvxpip install aimemory-mcp-server
pip- repohttps://github.com/jingchang0623-crypto/aimemory
- packagehttps://pypi.org/project/aimemory-mcp-server
- homepagehttps://aimemory.pro
- adoption0 stars · 0 forks
03Access granted
Query a database · destructive
The access this server can exercise, inferred from its verified tools — not a declared OAuth scope.
05Provenance & freshness
sourcesPyPI [p4]
last_checked2026-08-16 19:15Z
next_check2026-08-16 22:10Z
cadenceevery 3h
verifiedtools_list:passed handshake:passed metadata:passed tools_list:passed handshake:passed metadata:passed tools_list:passed handshake:passed metadata:passed tools_list:passed
index_statusindex — 8 unique facts >= 5
06Badge
Add the “as seen on MCPExplorer” badge to your README. 
[](https://mcpexplorer.com/servers/aimemory-mcp-server)
Next step
This is one server. A loadout combines the right servers, governance, and proven plays for a whole job — assembled deliberately, not tool-dumped.
Explore loadouts →