
Smara
Connects Claude (and other MCP clients) to the Smara Memory API, which ranks stored facts using a decay algorithm modeled on the Ebbinghaus forgetting curve. You get five tools: store_memory with importance weighting, search_memories with temporal scoring that surfaces recent and critical facts first, get_user_context for dropping formatted memory into system prompts, plus delete and usage checking. Requires a Smara API key. Reach for this when you want conversational context that persists across sessions and naturally deprioritizes stale information without manual pruning. Free tier includes 10,000 memories.
- tools
- 5
- auth
- Secret needed
- transport
- STDIO
- env var
- 1
- last updated
- Apr 6, 2026
Tools
5 toolsPublic tool metadata for what this MCP can expose to an agent.
store_memoryStore a fact about a user with importance scoring
Store a fact about a user with importance scoring
No parameter schema in public metadata yet.
search_memoriesSemantic search with Temporal Memory Scoring™
Semantic search with Temporal Memory Scoring™
No parameter schema in public metadata yet.
get_user_contextPre-formatted context string for LLM system prompts
Pre-formatted context string for LLM system prompts
No parameter schema in public metadata yet.
delete_memoryDelete a specific memory
Delete a specific memory
No parameter schema in public metadata yet.
get_usageCheck plan limits and memory count
Check plan limits and memory count
No parameter schema in public metadata yet.
Sponsors
Configuration
SMARA_API_KEY*secretYour Smara API key from smara.io
Readme
@smara/mcp-server
MCP server for the Smara Memory API — give any AI app persistent memory with Temporal Memory Scoring™.
Quick Start
Add to your Claude Desktop config (~/Library/Application Support/Claude/claude_desktop_config.json):
{
"mcpServers": {
"smara": {
"command": "npx",
"args": ["-y", "@smara/mcp-server"],
"env": {
"SMARA_API_KEY": "your-api-key-here"
}
}
}
}
Get a free API key at smara.io (10,000 memories, no credit card).
Tools
| Tool | Description |
|---|---|
store_memory | Store a fact about a user with importance scoring |
search_memories | Semantic search with Temporal Memory Scoring™ |
get_user_context | Pre-formatted context string for LLM system prompts |
delete_memory | Delete a specific memory |
get_usage | Check plan limits and memory count |
How It Works
Details
- Repository
- github.com/smara-io/mcp-server
- Category
- AI & LLM Tools
- Package
@smara/mcp-server- Transport
- STDIO
- Auth
- Required
- Registry
- MCP registry, active