# BanditDB

> BanditDB — com-banditdb-mcp. Use this tool when you need to optimize decision-making in dynamic environments, as BanditDB provides a persistent memory for agents to learn and recall effective actions in specific contexts. It solves problems of trial-and-error exploration and suboptimal decision-making by storing and retrieving successful actions. BanditDB takes in agent experiences and outputs informed decision recommendations, ideal for use in reinforcement learning and autonomous systems.

Canonical page: https://skillsregistry.net/skills/com-banditdb-mcp  
JSON: https://api.skillsregistry.net/v1/skills/com-banditdb-mcp

## Description

Persistent decision memory for agents — learns which action works in which context

## Trust

- **Trust score (0–1):** 0.70
- **Verification tier:** unverified

## Facts

- **Version:** 0.2.1
- **Skill type:** atomic
- **Execution layer:** mcp-remote
- **Runtime environment:** api
- **Updated:** 2026-09-02

## Source

- **Source listing:** [MCP Registry](https://registry.modelcontextprotocol.io/v0/servers/com.banditdb%2Fmcp)
- **Repository:** <https://github.com/dynamicpricing-ai/banditdb-python>

## Use it

Resolve this record through the SkillsRegistry MCP server (no auth, read-only):

```
claude mcp add --transport http --scope user skillsregistry https://api.skillsregistry.net/mcp
```

```json
{
  "jsonrpc": "2.0",
  "id": 1,
  "method": "tools/call",
  "params": {
    "name": "get_skill",
    "arguments": {
      "slug": "com-banditdb-mcp"
    }
  }
}
```

REST: `GET https://api.skillsregistry.net/v1/skills/com-banditdb-mcp` · pull for local use: `GET https://api.skillsregistry.net/v1/skills/com-banditdb-mcp/pull`

---
SkillsRegistry indexes agent skills from public registries and GitHub. Skills we have analysed are scanned with Circle-IR and scored on six dimensions; each listing states its scan coverage. More: https://skillsregistry.net/llms.txt