Bug-006 -: The nice-to-have skill count used during AI matching (5) does not match the actual nice-to-have count parsed from the JD (6) — one skill is silently dropped, skewing the match score.

Bug-006

Bug Summary:
The nice-to-have skill count used during AI matching (5) does not match the actual nice-to-have count parsed from the JD (6) — one skill is silently dropped, skewing the match score.

Environment: Staging
Module: AI Matching — Nice-to-have Skill Count

Preconditions
User is logged in.
JD "DevOps" parsed with 6 nice-to-have skills (Banking/regulated experience, Azure, GCP, Python, Shell scripting, Cloud cost optimization knowledge).
Candidate "Dinesh Inamdar" scored against this JD.

Steps to Reproduce
Login to the application.
Open the "DevOps" job's AI-parsed requirements page; confirm "Skills — 14 must-have · 6 nice-to-have".
Open Dinesh Inamdar's applicant detail page.
Check the "Nice-to-have" progress bar fraction and percentage under AI match.
Expand "Skill coverage" section and count Nice-to-have Matched + Missing chips.

Expected Result
Nice-to-have fraction should read 1/6 (not 1/5); Skill coverage should list all 6 nice-to-have skills across Matched + Missing, including "Cloud cost optimization knowledge" under Missing.

Actual Result
AI match shows "Nice-to-have 1/5 · 20%". Skill coverage shows Matched·1 + Missing·4 = 5 total. "Cloud cost optimization knowledge" does not appear anywhere. Recalculating the scoring formula confirms the Overall score (3.6/10) uses denominator 5, not 6 — using 6 would yield ~3.5/10.

Test Data
JD: DevOps (6 nice-to-have skills). Resume: Dinesh Inamdar (DINESH_INAMDAR_CV.pdf).

Severity
Medium

Priority
Medium-High

Attachnment :