claudebot-memory
v1.0.0
Published
MCP Server for persistent bot memory — stores learnings, preferences, skills and workflows in a local JSON store. Works with any MCP-compatible client (Claude Code, Claude Desktop, Cursor, etc.)
Maintainers
Readme
claudebot-memory
MCP Server for persistent AI memory. Stores learnings, preferences, skills and reusable workflows in a local JSON file.
Works with any MCP-compatible client: Claude Code, Claude Desktop, Cursor, Windsurf, etc.
Install
npx claudebot-memorySetup
Claude Code
claude mcp add claudebot-memory -- npx claudebot-memoryClaude Desktop
Add to claude_desktop_config.json:
{
"mcpServers": {
"claudebot-memory": {
"command": "npx",
"args": ["claudebot-memory"],
"env": {
"CLAUDEBOT_DATA_DIR": "/path/to/your/data"
}
}
}
}Cursor / Windsurf
Add to your MCP settings:
{
"claudebot-memory": {
"command": "npx",
"args": ["claudebot-memory"]
}
}Environment Variables
| Variable | Default | Description |
| -------------------- | -------- | ----------------------------------------- |
| CLAUDEBOT_DATA_DIR | ./data | Directory where memories.json is stored |
Tools
Memory Tools
| Tool | Description |
| --------------- | ------------------------------------------------ |
| memory_save | Save a learning (category, title, content, tags) |
| memory_search | Search learnings by keywords |
| memory_list | List all learnings by category |
| memory_delete | Delete an obsolete learning |
Categories: command, preference, contact, project, system, error, workflow
Skill Tools
| Tool | Description |
| --------------- | ------------------------------------------------------ |
| skill_create | Create a reusable skill with steps and prompt template |
| skill_search | Find a skill before starting a task |
| skill_list | List all available skills |
| skill_improve | Track success/failure and refine a skill after use |
Skills are reusable workflows with:
- Trigger: when to use the skill
- Steps: ordered actions to follow
- Prompt template: with
{{PLACEHOLDERS}}for variable parts - Success rate: automatically tracked over time
Data Format
All data is stored in memories.json:
{
"version": 2,
"memories": [
{
"id": "abc123",
"category": "command",
"title": "Deploy Next.js to Vercel",
"content": "Run: vercel --prod",
"tags": ["deploy", "vercel", "nextjs"],
"createdAt": "2025-01-01T00:00:00.000Z",
"updatedAt": "2025-01-01T00:00:00.000Z",
"usageCount": 5
}
],
"skills": [
{
"id": "def456",
"name": "create-api-route",
"description": "Create a Next.js API route handler",
"trigger": "When user asks to create an API endpoint",
"steps": [
"Create route.ts in app/api/",
"Add GET/POST handlers",
"Add Zod validation"
],
"prompt": "Create a Route Handler at app/api/{{RESOURCE}}/route.ts with {{METHOD}} method",
"tags": ["nextjs", "api"],
"usageCount": 3,
"successRate": 100,
"improvements": []
}
]
}Contributing
PRs welcome! This is part of ClaudeBot.
License
MIT
