Install once, remember everywhere. Claude Code, Cursor, Codex—they all share the same persistent memory. Auto-stores context silently. Ebbinghaus decay lets old facts fade while what matters persists.
Claude Code, Cursor, Windsurf—paste this and restart. Memory is automatic.
{
"smara": {
"command": "npx",
"args": ["-y", "@smara/mcp-server"],
"env": { "SMARA_API_KEY": "your-key-here" }
}
}
No SDKs to install. No schemas to define. Just REST.
curl -X POST https://api.smara.io/v1/memories \ -H "Authorization: Bearer smara_your_key" \ -H "Content-Type: application/json" \ -d '{ "user_id": "user_42", "fact": "Prefers dark mode. Uses vim keybindings.", "importance": 0.7 }'
{ "id": "mem_8f3a...", "decay_score": 1.0, "status": "stored" }
curl https://api.smara.io/v1/memories/search \ -H "Authorization: Bearer smara_your_key" \ -G \ -d user_id=user_42 \ -d q="editor preferences" \ -d limit=5
{ "memories": [
{ "fact": "Prefers dark mode. Uses vim keybindings.",
"similarity": 0.94, "decay_score": 0.97 }
] }
curl https://api.smara.io/v1/users/user_42/context \
-H "Authorization: Bearer smara_your_key"
{ "user_id": "user_42",
"memory_count": 23,
"context": [
{ "fact": "Prefers dark mode. Uses vim keybindings.", "score": 0.96 },
{ "fact": "Working on a Rust CLI tool for log parsing", "score": 0.89 }
] }
MCP for Claude Code, Cursor, Windsurf. OpenAI functions for Codex and ChatGPT. REST API for everything else. One memory pool, every tool.
Install the MCP server. That's it. Smara automatically loads context at conversation start and silently stores new facts as you work. No manual tool calls needed.
Ebbinghaus decay scoring means old trivia fades while important facts persist. Contradiction detection auto-replaces outdated memories. Just like human memory.
Every memory knows where it came from—Claude Code, Cursor, Codex, or your API. Cross-platform context with full provenance. Filter by source or see everything.
Free tier. 10,000 memories. No credit card.
Takes 2 seconds. Your key appears right here.