Banditkeep runs on your own machine — the client starts it, so there is no endpoint to ping. 270 installs a week from npm. Last commit 4 Nov 2026.
Decision memory for agents: bandit campaigns that learn which action works in which context.
Nothing serious here today
Today is the operative word: we check Banditkeep every 15 minutes and re-read its code on every release. Watch it and you find out the day that stops being true.
Connect this server
This server runs on your own machine — install it with the package manager and the client starts it for you. Package name taken from the official registry entry.
claude mcp add banditkeep -- npx -y @reneturcios/banditkeep{
"mcpServers": {
"banditkeep": {
"command": "npx",
"args": [
"-y",
"@reneturcios/banditkeep"
]
}
}
}[mcp_servers.banditkeep]
command = "npx"
args = ["-y", "@reneturcios/banditkeep"]{
"mcpServers": {
"banditkeep": {
"command": "npx",
"args": [
"-y",
"@reneturcios/banditkeep"
]
}
}
}{
"servers": {
"banditkeep": {
"type": "stdio",
"command": "npx",
"args": [
"-y",
"@reneturcios/banditkeep"
]
}
}
}Alternatives to Banditkeep
same job, measured the same wayPersistent decision memory for agents — learns which action works in which context
Git-native memory for coding agents: gotchas, decisions and bookmarks as plain text in your repo.
Memory for coding agents: the decisions, the dead ends, and where the last session stopped.
Local-first memory for coding agents; stores decisions, bugs, and context across sessions.
Session memory for AI coding agents. Search past sessions, extract rules, and track what works.
Memory that reasons: continual learning for stateful agents. Better context, fewer tokens.
Self-learning memory for AI agents, powered by the Ragionex context engine.
Persistent memory for AI coding agents — past bugs, decisions, and your corrections, in your repo.
Banditkeep — questions
Answers built from our own checks of this server.