fix: compare samplesheet enum values as strings in validator

[JOhnsonKC201]

Problem

In bio-research/skills/nextflow-development/scripts/utils/validators.py, the enum (allowed) check in validate_samplesheet compared raw CSV cell values against the configured allow-list with a plain in:

value = row[col_name]          # from CSV -> always a string, e.g. "0"
allowed = col_config["allowed"]  # may be typed, e.g. integers [0, 1]
if value not in allowed:         # "0" not in [0, 1] -> True

CSV values are read as strings, but enum allow-lists in the pipeline configs can be typed. The sarek config (scripts/config/pipelines/sarek.yaml) declares the status column as type: integer with allowed: [0, 1]. So the membership test "0" not in [0, 1] is always True, and the validator rejected every valid sarek samplesheet with errors like:

Row 2: Invalid value '0' for 'status'. Allowed: [0, 1]

Fix

Normalize both sides to strings before comparing, so typed enums validate correctly while still rejecting genuinely invalid values:

if str(value) not in [str(a) for a in allowed]:

Minimal, one-line behavioral change plus an explanatory comment.

Repro (manual — no test harness exists for these scripts)

Using a sarek-style integer enum column (status: allowed [0, 1]) and a valid samplesheet (status = "0" / "1"):

• Before: valid: False — ["Row 2: Invalid value '0' for 'status'. Allowed: [0, 1]", "Row 3: Invalid value '1' ..."]
• After: valid: True, no errors

Negative case still works: a genuinely invalid value "2" is still rejected after the fix. python -m py_compile on the changed file passes.

[bryan-anthropic]

Thanks for your interest! This repo only accepts contributions from Anthropic team members. If you'd like to submit a plugin to the marketplace, please submit your plugin here.