ldclabsldclabsOfficial·1 Agent Skills Included

anda-brain

Persistent long-term graph memory for language model applications

Encodes conversations into a structured knowledge graph, recalls past facts and preferences through natural language queries, and runs automatic maintenance to consolidate and prune memory. Eliminates stale vector search results, forgotten user preferences, and contradictory stored facts. Lets multiple applications share one evolving memory space through a simple REST API or MCP connection.
npx skills add ldclabs/anda-brain --all -g -y
Available:

Gives coding agents the repository layout, build and test commands, feature flags, and memory-system invariants they must follow when modifying this Rust service.

All Skills in This Repository (1)

Pure Emerald Level Indicators

Frequently Asked Questions

FAQPage Schema
How to install anda-brain?

Run `npx skills add ldclabs/anda-brain --all -g -y` in your terminal to install the skill globally.

What problem does Anda Brain solve?

It gives language model applications persistent long-term memory, so user preferences, facts, and past decisions survive across sessions instead of being forgotten or contradicted.

How is Anda Brain different from vector RAG?

It stores memory as a knowledge graph with timelines and supersession, so it can merge duplicates, mark outdated facts, and track how knowledge changes over time.

How do agents connect to Anda Brain?

Agents use a simple REST API or an MCP endpoint with natural language queries, with no need to learn graph query languages.

Can I self-host Anda Brain?

Yes. It is open-source and designed for self-hosting, with a quick start deployment guide and Docker support included.

Related Repositories in Software Engineering

View All in Software Engineering