ai-memory MCP
ai-memory is a persistent memory substrate for AI agents that provides encryption, cryptographic attestation, and a forensic audit trail. It ensures data integrity by guaranteeing deletes destroy encryption keys, maintains a tamper-evident audit chain, and supports exactly-once job claiming via atomic transitions. The server works with any MCP client and supports SQLite or PostgreSQL with Apache AGE and pgvector for scalable, secure memory storage across single agents to large swarms.
Package and access
Local package with source-attributed installation metadata. Koskamo does not execute packages during review.
Checks and signals
Each line is separate evidence, not a trust score.
- Popularity 56 GitHub stars
What you can do with it
- Provide persistent memory for AI agents across sessions
- Ensure data integrity with cryptographic attestation and audit trails
- Manage secure, scoped access for sub-agents using capability tokens
- Prevent race conditions in multi-agent environments with atomic claims
Source and maintenance
- Last repository push
- Latest release
- v0.10.0
- Maintenance checked
Packages and installation
Source-provided package data. Koskamo has not executed or live-probed this package.
ghcr.io/alphaonedev/ai-memory:0.5.1
{
"transport": "stdio",
"identifier": "ghcr.io/alphaonedev/ai-memory:0.5.1",
"runtimeHint": "docker",
"registryType": "oci",
"packageArguments": [
{
"type": "positional",
"value": "mcp"
},
{
"name": "--tier",
"type": "named",
"value": "semantic",
"description": "Feature tier: keyword, semantic, smart, autonomous"
}
],
"environmentVariables": [
{
"name": "AI_MEMORY_DB",
"description": "Path to the SQLite database file"
}
]
}Tools
Reported by the attributed source; not independently observed by Koskamo.
Maintenance history
Public check and source-refresh events. Editorial and ownership records remain private.
- · repository · passed · metadata changed · updated
- · backfill · recorded · Pre-expansion baseline; unobserved rich metadata remains unknown
Connect from your client
Each snippet configures the same source-provided package; replace placeholders before use.
Cursor
{
"mcpServers": {
"ai-memory-mcp": {
"command": "docker",
"args": [
"run",
"-i",
"--rm",
"-e",
"AI_MEMORY_DB=<YOUR_AI_MEMORY_DB>",
"ghcr.io/alphaonedev/ai-memory:0.5.1",
"mcp",
"--tier=semantic"
],
"env": {
"AI_MEMORY_DB": "<YOUR_AI_MEMORY_DB>"
}
}
}
}Cline
{
"mcpServers": {
"ai-memory-mcp": {
"command": "docker",
"args": [
"run",
"-i",
"--rm",
"-e",
"AI_MEMORY_DB=<YOUR_AI_MEMORY_DB>",
"ghcr.io/alphaonedev/ai-memory:0.5.1",
"mcp",
"--tier=semantic"
],
"env": {
"AI_MEMORY_DB": "<YOUR_AI_MEMORY_DB>"
}
}
}
}VS Code
{
"servers": {
"ai-memory-mcp": {
"type": "stdio",
"command": "docker",
"args": [
"run",
"-i",
"--rm",
"-e",
"AI_MEMORY_DB=<YOUR_AI_MEMORY_DB>",
"ghcr.io/alphaonedev/ai-memory:0.5.1",
"mcp",
"--tier=semantic"
],
"env": {
"AI_MEMORY_DB": "<YOUR_AI_MEMORY_DB>"
}
}
}
}Claude Desktop
{
"mcpServers": {
"ai-memory-mcp": {
"command": "docker",
"args": [
"run",
"-i",
"--rm",
"-e",
"AI_MEMORY_DB=<YOUR_AI_MEMORY_DB>",
"ghcr.io/alphaonedev/ai-memory:0.5.1",
"mcp",
"--tier=semantic"
],
"env": {
"AI_MEMORY_DB": "<YOUR_AI_MEMORY_DB>"
}
}
}
}Let your agent find its tools
Find servers like ai-memory MCP from your agent. Add Koskamo once to search the index for endpoints, access requirements, and client setup. Public discovery needs no key.