ScottRBKScottRBKCommunityยท10 Agent Skills Included

forgetful

Shared memory and knowledge graph for coding workflows

Stores decisions, solutions, and preferences as atomic memories that any connected coding client can search semantically. Automatically links related memories into a knowledge graph, eliminating lost context between sessions and repeated explanations. Manages projects, entities, documents, code snippets, procedures, and files in one place with local SQLite or PostgreSQL storage.
npx skills add ScottRBK/forgetful --all -g -y
Available:

Gives the agent the project's architecture rules, testing approach, and linting commands to follow when working in this repository.

All Skills in This Repository (10)

Pure Emerald Level Indicators

Frequently Asked Questions

FAQPage Schema
How to install Forgetful?โ–ผ

Run `npx skills add ScottRBK/forgetful --all -g -y` in your terminal to install all skills in this suite globally.

What does Forgetful do?โ–ผ

It is a shared memory server that lets your coding clients store and recall decisions, solutions, and project knowledge across sessions. Memories are embedded semantically and auto-linked into a knowledge graph.

How do I connect Claude Code or Cursor to Forgetful?โ–ผ

Add it as an MCP server with the stdio command `uvx forgetful-ai`, or point your client at a running HTTP endpoint. The forgetful-mcp-setup skill walks through verification for each client.

Does Forgetful need a cloud database or API keys?โ–ผ

No. It runs locally with zero-config SQLite by default, and embeddings are generated offline with FastEmbed. PostgreSQL is optional for shared or production deployments.

Can I use Forgetful from the terminal without an MCP client?โ–ผ

Yes. The `forgetful` CLI exposes the same operations, with commands like `forgetful memory save` and `forgetful memory search`, plus `--json` output for scripting.

Related Repositories in Software Engineering

View All in Software Engineeringโ†’