Check minknow experiment folder on startup

[berndbohmeier]

We checked that the minknow experiment folder is created only a few sec after the minknow experiment is started, probably after some checks are run. We decided this is good enough to check it so that people who misspelled something will be notified. Some folders inside are only created much later, so we should not wait for that.

[Copilot]

🟡 Changes recommended

Four moderate findings remain unresolved.

Once you've addressed the issues Copilot identified, you can request another Copilot review.

Pull request overview

Adds startup validation for MinKNOW experiment folders and improved typo suggestions for experiment and metadata names.

Changes:

• Validates experiment directories during path resolution.
• Adds close-match suggestions for experiments and metadata files.
• Routes input errors through user-facing CLI messages.

File summaries

File	Summary and findings
src/nomadic/util/minknow.py	Adds experiment validation and similarity matching. Moderate findings: use is_dir(), exclude reserved directories, and normalize case-insensitive comparisons (3, 3, and 1 votes).
src/nomadic/realtime/commands.py	Adds metadata matching and error handling. Moderate finding: restrict candidates to .csv and .xlsx files (2 votes).

Review details

Suppressed comments (1)

src/nomadic/util/minknow.py:140

• The query is lowercased but the candidate directory names are not, so the supposed close-match lookup is still case-sensitive. For an uppercase MinKNOW experiment such as ABCDEF, even a one-character typo can fall through to the generic message; normalize both sides for comparison while returning the original directory name.

    close_matches = difflib.get_close_matches(
        experiment_name.lower(),
        [d.name for d in minknow_path.glob("*") if d.is_dir()],
        cutoff=0.8,

• Files reviewed: 2/2 changed files
• Comments generated: 3
• Review effort level: Lite

---

💡 Add a code-review agent skill or configure MCP servers for context-aware, tailored reviews. Learn more in the docs.