yertle-python

Python ecosystem for Yertle — CLI, SDK, and tooling for working with Yertle architecture graphs from Python.

Install

The CLI and SRE agent are distributed on PyPI and installed with uv:

uv tool install "yertle[sre,mcp]"

That puts three commands on your PATH:

Command	What it does
yertle	CLI — version, login, orgs list, nodes list, nodes tree, auth status
yertle-sre	Natural-language SRE agent
yertle-mcp	MCP server for AI agents

Pick fewer extras if you want fewer commands — uv tool install "yertle[cli]" installs just yertle. pipx install "yertle[sre,mcp]" works too.

Verify:

yertle version
yertle auth status     # shows which API URL and token are in effect

Using it as a library

For the SDK alone, install into your project rather than as a tool:

pip install yertle          # or: uv add yertle

import yertle

yertle.configure(token="yrt_...")
for org in yertle.orgs.list():
    print(org.name)

Credentials resolve from $YERTLE_TOKEN then ~/.yertle/config.json; the API URL from $YERTLE_API_URL, then the config file, then https://api.yertle.com. yertle auth status reports which source won for each.

Development

Requires uv and Python 3.11+.

make install        # uv sync --extra cli --extra dev
make check          # lint + format-check + typecheck + test

See make help for the full target list.

License

MIT. See LICENSE.