Back to Browse
Free
Server data from the Official MCP Registry
Ebbinghaus-based persistent memory for Claude. Memories decay with time, strengthen on recall.
About
Ebbinghaus-based persistent memory for Claude. Memories decay with time, strengthen on recall.
What You'll Need
Set these up before or after installing:
PostgreSQL connection string with pgvector. Example: postgresql://localhost:5432/yourmemory
Environment variable: DATABASE_URL
Ollama server URL for local embeddings (nomic-embed-text) and classification (llama3.2:3b). Default: http://localhost:11434
Environment variable: OLLAMA_URL
Ollama model used for fact/assumption classification. Default: llama3.2:3b
Environment variable: EXTRACT_MODEL
How to Install
Add this to your MCP configuration file:
{
"mcpServers": {
"io-github-sachitrafa-cognitive-ai-memory": {
"env": {
"OLLAMA_URL": "your-ollama-url-here",
"DATABASE_URL": "your-database-url-here",
"EXTRACT_MODEL": "your-extract-model-here"
},
"args": [
"-y",
"github:sachitrafa/cognitive-ai-memory"
],
"command": "npx"
}
}
}Security Report
5.5
Moderate5.5Moderate RiskWell-structured MCP server with strong security practices. Uses environment variables for database configuration, implements proper PostgreSQL connections, and includes reasonable access controls. Minor issues around broad exception handling and potential SQL injection in one query, but overall demonstrates good security hygiene. Supply chain analysis found 3 known vulnerabilities in dependencies (1 critical, 2 high severity).
Scanned 7 files · 6 findings
Security scores are indicators to help you make informed decisions, not guarantees. Always review permissions before connecting any MCP server.
Permissions Required
This plugin requests these system permissions. Most are normal for its category.
Reviews
No reviews yet
Be the first to review this server!