[Quality] Review top 10 hypothesis pages for demo quality

← All Specs

Goal

Review the 10 highest-scored hypothesis pages for demo quality. Check title quality, description completeness, evidence tables, radar chart rendering, related hypotheses linking, and challenge connections. Fix any rendering or data issues found.

Acceptance Criteria

☐ All 10 top hypothesis pages return HTTP 200
☐ Evidence sections show populated cards (not empty)
☐ Radar charts display real dimension score data
☐ Related hypotheses section present with links
☐ Challenge connections visible
☐ Paper titles display correctly (not just PMID:xxxxx)
☐ Tab counts show accurate badge numbers

Approach

  • Get top 10 hypotheses by composite score via API
  • Check each page for HTTP 200, section presence, data completeness
  • Identify and fix rendering bugs
  • Verify fixes with syntax check
  • Dependencies

    • Live API at localhost:8000

    Work Log

    2026-04-07 — Task 9b4b1e14 (slot 22)

    Review findings:

    All 10 pages return 200 OK. Key quality checks passed:

    • Titles: present, informative (though browser tab truncated to 60 chars)
    • Evidence: 37-44 supporting, 13-19 opposing per page, rendered as styled cards
    • Radar charts: real dimension score data (0.1-0.95 range)
    • Related hypotheses: 8 hypotheses from same analysis linked
    • Challenge connections: challenge link + bounty amount displayed
    • Price history: SVG charts present
    • Notebooks: 1 linked per variant hypothesis
    Bugs fixed:

  • Papers tab: missing titlesLIMIT 30 in paper metadata query cut off older papers
  • (sorted by year DESC), so their entries were absent from paper_map and rendered as
    "PMID:XXXXX" with no title. Fixed by using LIMIT {max(len(all_pmids), 30)} to fetch
    all needed paper records.

  • Papers tab: no count badge — The Papers tab button had no evidence count badge
  • unlike all other tabs. Added linked_paper_count variable and badge to tab button.

  • Papers tab: no header — Added "Cited Papers (N)" heading to the papers panel
  • for consistency with other panels (which all have headers).

  • Title tag too short<title> tag truncated hypothesis title at 60 chars,
  • meaning browser tabs always showed cut-off titles. Increased to 120 chars.

    File: quality_top10_hyp_pages_spec.md
    Modified: 2026-04-28 03:24
    Size: 2.3 KB