Skip to main content

​
RAG

RAG provides readonly access to databases and filesystems for production safety. 
from a1 import RAG, Agent, LLM

# Readonly database access
rag = RAG(database="sqlite:///data.db")

# Readonly filesystem access
rag = RAG(filesystem="/path/to/files")

agent = Agent(
    name="safe_agent",
    tools=[rag.get_toolset(), LLM("gpt-4.1")]
)
Database tools: sql (SELECT queries only) Filesystem tools: ls (list files), grep (search), cat (read)

​
Database

Full database access with read and write operations. 
from a1 import Database, Agent, LLM

db = Database("postgresql://user:pass@host/db")

agent = Agent(
    name="db_agent",
    tools=[db.get_toolset(), LLM("gpt-4.1")]
)
Available tools: sql (any query), insert, update, delete Supported databases: PostgreSQL, MySQL, SQLite, DuckDB, SQL Server, Oracle (any SQLAlchemy connection string)

​
FileSystem

Full filesystem access with read and write operations. 
from a1 import FileSystem, Agent, LLM

# Local filesystem
fs = FileSystem("/data")

# S3 bucket
fs = FileSystem("s3://bucket/path")

# Google Cloud Storage
fs = FileSystem("gs://bucket/path")

agent = Agent(
    name="fs_agent",
    tools=[fs.get_toolset(), LLM("gpt-4.1")]
)
Available tools: ls (list), grep (search), cat (read), write_file, delete_file Supported paths: Local filesystems, S3, Google Cloud Storage, Azure Blob Storage