personalrag
Núcleo reutilizável de RAG (Retrieval-Augmented Generation) sobre bases de conhecimento em Markdown.
Separa o motor (esta biblioteca) do conteúdo/config de cada projeto. Cada base define um Settings e usa a API:
from personalrag import VaultRag, Settings
settings = Settings(vault_dir="..", collection="git")
rag = VaultRag(settings)
rag.index()
print(rag.buscar("como desfazer o último commit"))
print(rag.responder("como desfazer o último commit"))
E o servidor MCP:
from personalrag.mcp import run_mcp
from config import settings
run_mcp(settings)
Instalação
pip install personalrag # núcleo
pip install "personalrag[mcp]" # com o servidor MCP
Módulos
| Módulo | Responsabilidade |
|---|---|
config.py |
Settings (dataclass) |
chunking.py |
Dividir markdown por seção + breadcrumb |
store.py |
Embeddings + Chroma |
retrieve.py |
Busca (+ rerank opcional) |
generate.py |
LLM compatível com OpenAI |
rag.py |
Classe VaultRag |
cli.py |
python -m personalrag index/query |
mcp.py |
run_mcp(settings) |
1 maintainer
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
Source Distribution
personalrag-0.1.0.tar.gz
(8.7 kB
view details)
Built Distribution
Filter files by name, interpreter, ABI, and platform.
If you're not sure about the file name format, learn more about wheel file names.
Copy a direct link to the current filters
File details
Details for the file personalrag-0.1.0.tar.gz.
File metadata
- Download URL: personalrag-0.1.0.tar.gz
- Upload date:
- Size: 8.7 kB
- Tags: Source
- Uploaded using Trusted Publishing? No
- Uploaded via:
twine/6.2.0 CPython/3.11.2
File hashes
| Algorithm | Hash digest | |
|---|---|---|
| SHA256 |
957c112999bdba9554352e63f9511e805e47bff88fcf05f2c540e294ee5c106f
|
|
| MD5 |
b5b267b8ff90986a42f5f8ee08a5dfa4
|
|
| BLAKE2b-256 |
42427740ef2e54e33851174a03ed3ae70937581c2c45aaeb5d6eff1e13e10ddb
|
File details
Details for the file personalrag-0.1.0-py3-none-any.whl.
File metadata
- Download URL: personalrag-0.1.0-py3-none-any.whl
- Upload date:
- Size: 10.2 kB
- Tags: Python 3
- Uploaded using Trusted Publishing? No
- Uploaded via:
twine/6.2.0 CPython/3.11.2
File hashes
| Algorithm | Hash digest | |
|---|---|---|
| SHA256 |
c625272fea226a4cf66e948e377ca46d343fef38e2fb547778c92563b36460f2
|
|
| MD5 |
dc8f42b71cec8983025ab69964429ede
|
|
| BLAKE2b-256 |
29a29ce8741aa319307c3735c3072b8f8e799b9ba3108484223da002ac17aa5d
|
