Governed memory for long-lived coding agents with trust levels, provenance and conflict detection.
Setup
Choose your client, copy the config, then fill in any required environment variables.
claude mcp add agentmem -- uvx quilmem --db '<DB>' --project '<PROJECT>'~/.codex/config.toml
[mcp_servers.agentmem]
command = "uvx"
args = ["quilmem", "--db", "<DB>", "--project", "<PROJECT>"]
# codex mcp add agentmem -- uvx quilmem --db '<DB>' --project '<PROJECT>'~/.cursor/mcp.json
{
"mcpServers": {
"agentmem": {
"command": "uvx",
"args": [
"quilmem",
"--db",
"<DB>",
"--project",
"<PROJECT>"
]
}
}
}.vscode/mcp.json
{
"servers": {
"agentmem": {
"type": "stdio",
"command": "uvx",
"args": [
"quilmem",
"--db",
"<DB>",
"--project",
"<PROJECT>"
]
}
}
}~/.gemini/settings.json
{
"mcpServers": {
"agentmem": {
"command": "uvx",
"args": [
"quilmem",
"--db",
"<DB>",
"--project",
"<PROJECT>"
]
}
}
}claude_desktop_config.json
{
"mcpServers": {
"agentmem": {
"command": "uvx",
"args": [
"quilmem",
"--db",
"<DB>",
"--project",
"<PROJECT>"
]
}
}
}.trae/mcp.json
{
"mcpServers": {
"agentmem": {
"command": "uvx",
"args": [
"quilmem",
"--db",
"<DB>",
"--project",
"<PROJECT>"
]
}
}
}