roampal-core
Active·★ 54·Apache-2.0·Updated 2026-09-26
Roampal Core provides outcome-based memory for AI coding assistants, enabling them to learn and improve their advice across multiple sessions. It automatically injects relevant context into conversations, promoting good advice and demoting bad advice to make the AI smarter over time.
#AI Memory#LLM#RAG#Python#Coding Assistant#Outcome-Based Learning#Context Injection#Developer Tool
01
Features
01Implements an outcome-based memory system that learns from good and bad advice, making the AI smarter over time.
02Automatically and seamlessly injects relevant context into AI coding assistant conversations without manual intervention.
03Utilizes five distinct memory collections (working, history, patterns, memory_bank, books) to ensure persistent learning across sessions.
04Allows ingestion of external reference documents ('books' collection) that are permanently searchable by the AI.
05Compatible with popular AI coding tools such as Claude Code and OpenCode, with platform-specific integration mechanisms.
02
Compatibility
Python
Runtime
Verified via docs
Claude Code
AI Tool
Verified via docs
OpenCode
AI Tool
Verified via docs
Windows
OS
Verified via docs
macOS
OS
Verified via docs
Linux
OS
Verified via docs
03
Quick start
1
$ pip install roampal
2
$ roampal init
04
Use cases
↳Enhancing AI coding assistants to remember architecture, preferences, and conventions across different sessions.
↳Improving the quality of AI-generated code and suggestions by learning from past outcomes (promoting good, demoting bad).
↳Providing AI with a searchable personal knowledge base from uploaded documentation for development tasks.
↳Reducing the need to repeat context to AI coding tools, enabling more efficient and productive coding sessions.
05
Alternatives
sparql-llm★ 130
🦜✨ Chat system, MCP server, and reusable components to improve LLMs capabilities when generating SPARQL queries
ragflow★ 92.6k
RAGFlow is a leading open-source Retrieval-Augmented Generation (RAG) engine that fuses cutting-edge RAG with Agent capabilities to create a superior context layer for LLMs
brain-mcp★ 80
Your AI has amnesia. Persistent memory and cognitive context for AI. 25 MCP tools. 12ms recall.
rekal★ 53
Long-term memory for LLMs. MCP server backed by hybrid search in a single SQLite file.
Vault-for-LLM★ 50
🧠 Local-first knowledge system for LLM agents — sqlite-vec + ONNX embeddings, no cloud/Docker/PyTorch dependency
GPT Researcher★ 30.2k
An autonomous agent that conducts deep research on any data using any LLM provider, producing detailed reports with citations from web and local sources.
Comments
Log in to leave a comment
No comments yet. Be the first!