[DevSecNinja]

Summary

• Deploy ~/Taskfile.yml and ~/.config/mlx/chat.py through Chezmoi so task -g llm:chat works from any directory in Fish, Bash, and Zsh; leave the repository Taskfile unchanged.
• Use Apple's mlx-lm API with idempotent native Apple Silicon setup in ~/mlx-qwen/.venv, mounted-SSD checks before setup, and Hugging Face/model/setup caches on the external SSD only.
• Run non-thinking Qwen3.5 chat with an 8,192-token admission limit, 512-token replies/prefill, reusable in-memory history, reset/quit/interrupt handling, and no transcripts.
• Document narrow deployment, including creation of the launcher's parent directory before selected-file Chezmoi apply.

The existing user-tested implementation was imported unchanged from the previous session. The user has confirmed live deployment works and authorized merging.

Validation

• mise install completed without upgrading existing tools.
• All-files and staged lefthook pre-commit checks passed; commit-msg verification passed.
• All 19 focused Bats checks passed, including 24 model-free Python tests and fresh-home selective deployment.
• Isolated Chezmoi initialization and chezmoi apply --dry-run --source=. passed. No live home apply, model inference, model download, or external cache edits were performed.
• The complete local Bats CI run executed all 895 tests. It reported 29 unrelated macOS/platform/environment failures (Linux-only command paths and GNU flags, missing timeout, terminal/installed-tool assumptions). 28 reproduced on unchanged main; the remaining existing macup test discovers the installed mas CLI and was not repeated to avoid live application upgrades. No unrelated fixes are included.
• Required GitHub CI checks must pass before merge; no hook or branch-protection bypass.

[coderabbitai]

Warning

Review limit reached

You've used all free OSS reviews for now. Wait for the free limit to reset to keep reviewing this public repository.

Next included review available in 55 minutes.

Check out review usage here.

View limit details

Limit details: You’ve used the included review currently available.

Learn how review limits work.

Review configuration:

⚙️ Run configuration

• Configuration used: defaults
• Review profile: CHILL
• Plan: Advanced
• Run ID: ccf0edd0-139e-4582-bcbe-9b502fd26af5

📥 Commits

Reviewing files that changed from the base of the PR and between cadaf68 and 23e3f39.

📒 Files selected for processing: 7

• README.md
• docs/development-tools.md
• home/.chezmoiignore
• home/Taskfile.yml
• home/dot_config/mlx/chat.py
• tests/bash/global-mlx-chat.bats
• tests/python/test_mlx_chat.py

• Autofix · Keep fixing CodeRabbit findings and required CI, and resolving merge conflicts

Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out.

❤️ Share

• X
• Mastodon
• Reddit
• LinkedIn

_{Comment @coderabbitai help to get the list of available commands.}

[DevSecNinja]

Superseded by #923, the canonical PR for this feature. Its launcher and documentation are identical, and it includes the verified test-only fix to resolve mise shims before switching fixture HOME. Closing this duplicate; only #923 will be merged.