Agentic ETL for Python. Declare pipelines in YAML where each step is a connector, a function or an LLM agent. Read CSV, Excel, PDF, Word, email, HTTP, S3 and SQL, validate outputs with schemas, and get a live diagram, local dashboard and notifications.
Setup
Choose your client, copy the config, then fill in any required environment variables.
claude mcp add agentl -- uvx agentl-etl~/.codex/config.toml
[mcp_servers.agentl]
command = "uvx"
args = ["agentl-etl"]
# codex mcp add agentl -- uvx agentl-etl~/.cursor/mcp.json
{
"mcpServers": {
"agentl": {
"command": "uvx",
"args": [
"agentl-etl"
]
}
}
}.vscode/mcp.json
{
"servers": {
"agentl": {
"type": "stdio",
"command": "uvx",
"args": [
"agentl-etl"
]
}
}
}~/.gemini/settings.json
{
"mcpServers": {
"agentl": {
"command": "uvx",
"args": [
"agentl-etl"
]
}
}
}claude_desktop_config.json
{
"mcpServers": {
"agentl": {
"command": "uvx",
"args": [
"agentl-etl"
]
}
}
}.trae/mcp.json
{
"mcpServers": {
"agentl": {
"command": "uvx",
"args": [
"agentl-etl"
]
}
}
}