The release that made pip install pengy mean something on a machine with no API key.
The default endpoint is auto-detected. Pengy used to default to api.openai.com. On a machine
with no OpenAI key that produces a 401 you have to fix before anything works, which is a poor first
experience for software whose whole premise is running against a model you own. It now probes
localhost for ollama and, failing that, OpenAI's localhost default, and uses whichever answers. The
endpoint switch is a single /endpoint slash command.
The CLI takes a prompt as an argument. pengy-cli "summarise this log" works. Before this it
needed a flag. Skills are wired into the CLI with /skills.
The rest is about not crashing. The startup fix chain in this project is long, and this is its
biggest instalment: a missing Qt, a missing PySide6, a missing assets/ in the wheel, an
incompatible PySide6 on Python 3.13, all of which produced "pengy-gui is not installed" followed by a
traceback. A missing-Qt install hint, a pengy-gui entry point that imports lazily, a
pengy doctor command, and assets/ shipped in the wheel. doctor is the answer to "I installed it
and it doesn't work", and it exists because that happens a lot.
1.9.0 (17 September) is the same day's work in the native editions, and the theme of it is
coherence: remote sudo over ssh with the password through a single-use askpass helper on the remote
side, the C++ web UI rebuilt on a real HTTP server after its prototype collapsed, the web worker
refactored into an embeddable class in all three editions, and pengy-cli as a real entry point in
Rust with --list-models and /endpoint.