F4 (Low): frontend lint 537 errors (532 auto-fixable), not in CI — pending #154 #179

What was found

ChatGPT-generated testing report at commit c7bebd1 ran npm run lint and reported:

• 537 errors
• 2 warnings
• 532 of 537 marked auto-fixable (Prettier formatting)
• Frontend tests pass (147 passing)
• Frontend production build passes
• Quality-gate workflow does NOT run frontend test/build/lint commands

The 5 non-auto-fixable errors were not enumerated in the report.

Status: pending #154

The current Lovable-generated frontend is under reconsideration (#154 Streamlit spike, areas/frontend-reconsideration-2026-11-02.md). No cleanup action is being taken for this issue today because the lint baseline is a frontend-only artefact that goes away with any replacement:

• Lovable frontend retained → fix lint baseline (npm run lint -- --fix for the 532 auto-fixables, investigate the 5 others), then add frontend checks to CI
• Streamlit replacement → Lovable code is deleted; this lint baseline is deleted with it
• Different framework → re-evaluate when that direction is picked

Consistent with #176 (RV-F1 pending #154) and the general stance on csv-table-hub-main/* code.

Caveat

The 5 non-auto-fixable errors could be substantive (unused variables, missing effect dependencies, actual bugs). If any are, deferring hides them until #154 settles. If #154 stays open for weeks and this concern grows, triage by running npm run lint 2>&1 | grep -v "prettier/prettier" to list only the non-formatting errors — a 15-minute exercise that does not block.

Related to G62 / RV-D4

RV-D4 is the broader architectural gap "CI completeness. Frontend checks not currently in CI." This issue is one concrete sub-item under it: frontend lint specifically. Fixing this contributes to RV-D4 but does not close it (tests, build, type-check also need adding).

Done when

• Spike: rebuild M4 review screen in Streamlit (1-day time-box) #154 has settled on a frontend direction.
• If Lovable retained: lint baseline resolved and npm run lint added to the quality-gate workflow.
• If Lovable replaced: this issue closes as moot.

Reference

Tracker row G64 (main tab). ChatGPT testing report F4, 3 Nov 2026. Related: #154 (frontend reconsideration), #176 (RV-F1 same pending-#154 stance), G62/RV-D4 (CI completeness, parent architectural item).