Skip to content

fix(ui): label dashboard figures for what they measure - #4368

Merged
swati510 merged 2 commits into
mainfrom
fix/dashboard-label-truth
Oct 22, 2026
Merged

swati510 merged 2 commits into
mainfrom
fix/dashboard-label-truth

Conversation

@RaghavChamadiya

Copy link
Copy Markdown
Member

What

Three dashboard figures said something their data did not measure.

  • Workspace "Doc freshness" averaged wiki page confidence (doc_coverage_pct). Confidence records how a page was written (template 1.0, model 0.8); it does not say whether the page is current.
  • Workspace system map health. A repo that health had not scored got a stand-in "health" score: 60% mean page confidence plus 40% hotspot share. The map drew it as a code health mark (for example "9.3 Healthy").
  • Overview code health door and inventory row said "11,371 open findings". get_overview, repowise health and the same payload's open_findings all say 13,563. The UI read the queue inventory, which leaves out zero-impact findings: 1,401 performance and 791 governance, 2,192 in total.

How

  • /api/workspace repo entries carry fresh_page_count. The pages, fresh pages and confidence come from one query, so each repo costs 6 sqlite statements instead of 7, and the pinned budget is updated to match. The workspace ribbon is fresh pages over all pages, floored, which is the share the single-repo overview reports. Repo rows also get "% of N doc pages fresh".
  • /api/workspace/graph sends health_score: null when health has not scored a repo, and the stand-in score and health_score_source are removed. The system map leaves such repos out of its health join, so their nodes read "No health score".
  • RepoCard reads "N% doc confidence" instead of "N% docs". WorkspaceGraphNode reads "N% doc conf." instead of "N% cov".
  • inventoryRows and buildExplore read health.open_findings.
  • en and zh-CN strings, regenerated HTTP types, and one line each in WORKSPACES.md and COMPONENT_CONTRACTS.md.

repowise health was already on the same definition (open_finding_count): 13,563 on the same index. No CLI change.

Verification

repowise serve on a copy of this repository's index, set up as a one-repo workspace, with curl of the routes and the overview model functions run on the served payload:

Figure Before After
Workspace Doc freshness 99%, "weighted across 7,253 pages" 86%, "6,263 of 7,253 pages fresh"
Single-repo overview freshness (unchanged) 86.4% 86.4%
Graph health for an unscored repo stand-in composite (93 for this repo's figures) null, "No health score"
Graph health for this repo (scored) 79.9 79.9
Overview code health door 11,371 open findings 13,563 open findings
Inventory health row 11,371 open findings 13,563 open findings
get_overview open_findings / repowise health 13,563 / 13,563 13,563 / 13,563

Tests: workspace router (fresh pages, null health, query budget), generated HTTP types, overview and workspace pytest suites (371 passed); ui vitest (2,781 passed); new web test for the freshness helper and both locales. Type-check is clean for ui, web and api-client. health --score --against origin/main on the touched files shows no drop.

@repowise-bot

repowise-bot Bot commented Oct 21, 2026 •

Copy link
Copy Markdown
Contributor

Reviewed the whole diff: nothing to fix.

✅ Health of changed files: 8.38 → 8.39 (+0.01; structure +0.01)
Average full file score (structure and history) of 16 changed code files, weighted by size before this PR; not the repository score.

Check Result
Coverage 100% of 6 changed lines covered
Tests 4 of 12 changed code files reached by a test · 7 reached by no test · 1 with no test signal · which tests
Security No new findings on changed lines
Doc drift 10 refuted claims in a doc this PR edits · 2 documents name changed files
Coverage

Patch coverage 100.0% (6 of 6 changed executable lines covered)

b1713a8 · cobertura · 2 of 20 changed files measured, 18 out of scope

Report from this run for b1713a8.

Doc drift
  • packages/ui/COMPONENT_CONTRACTS.md line 164 refers to DecisionHealthWidget, which the tree no longer has (high confidence)
  • packages/ui/COMPONENT_CONTRACTS.md line 568 refers to SymbolGraphPanel, which the tree no longer has (high confidence)
  • packages/ui/COMPONENT_CONTRACTS.md line 594 refers to SortCol, which the tree no longer has (high confidence)
  • +9 more on the PR page
🧾 Evidence: 6 not known
Item Status Evidence
Components touched ok · inferred 3 components: docs, packages, tests
Declared owners ok · declared @RaghavChamadiya, @swati510
Suggested reviewers unknown · not known Not known: not gathered on a pull request
Tests reaching the change unknown · not known 0 of 12 changed files reached by a test; 12 not known
Contracts touched ok · measured No workspace contract touched; no breaking signature change
Consumers in other repos unknown · not known Not known: not gathered on a pull request
Decisions crossed unknown · not known Not known: decision records are read only for an AI review
Health delta ok · measured Findings: 0 introduced, 0 worsened, 0 resolved
AI provenance unknown · not known Provenance: unknown, known for 0 of 2 commits; AI share withheld: coverage 0% is below 30%
Diff size rank ok · measured Larger than 49.6% of recent commits
Outcomes since merge unknown · not known Not known: a pull request is not a merged change

📊 See the full report for this PR
Blast radius, every caller of the contracts it changes, and health before and after. No sign-in.
Reviewed with AI · Plain markdown for agents

Settings and updates

Updated 2026-10-22 07:46 UTC
Silence one PR with [skip repowise] in the title · Bot settings
⭐ Star Repowise · 📥 Install on another repo

@RaghavChamadiya
RaghavChamadiya force-pushed the fix/dashboard-label-truth branch from 1032c7b to 3c9d402 Compare October 22, 2026 04:31
The workspace "Doc freshness" figure averaged page confidence; it is now
fresh pages over all pages, from a fresh_page_count on each workspace repo
entry. The repo card and graph node name that confidence figure as doc
confidence instead of docs or coverage.

A workspace repo that health has not scored had a stand-in score built from
doc confidence and hotspot share, drawn on the system map as code health. The
graph sends null for it instead, and the map reads "No health score".

The Overview's code health door and inventory row print open_findings, the
total get_overview and `repowise health` print, instead of the queue
inventory, which leaves zero-impact findings out.
@RaghavChamadiya
RaghavChamadiya force-pushed the fix/dashboard-label-truth branch from 3c9d402 to b1713a8 Compare October 22, 2026 07:44
@swati510
swati510 merged commit 323923a into main Oct 22, 2026
10 checks passed
@swati510
swati510 deleted the fix/dashboard-label-truth branch October 22, 2026 13:23
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants