Skip to content

Conversation

@codeaholicguy
Copy link
Owner

Problem

AI assistants lose persistent, actionable guidance across tasks. Users must repeatedly re-inject rules and conventions, leading to inconsistency and wasted effort.

Solution

Provide a lightweight knowledge memory service (SQLite + FTS5, MCP-compatible) plus a fallback CLI (ai-devkit memory store/search) and updated command templates that instruct agents to query/store knowledge or use the CLI when MCP is unavailable.

MCP Setup Example

Add this MCP server entry to your client config:

{
  "mcpServers": {
    "ai-devkit-memory": {
      "command": "npx",
      "args": ["@ai-devkit/memory"]
    }
  }
}

CLI

Add a user-facing ai-devkit memory CLI and update command templates so users can store/search reusable knowledge even when MCP tools aren’t available. Clean up the memory package to run as a dedicated MCP server.

@codeaholicguy codeaholicguy merged commit 5fe368a into main Jan 22, 2026
1 check passed
@codeaholicguy codeaholicguy deleted the feature/knowledge-memory branch January 22, 2026 15:03
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants