fix(observability): every Grafana query-variable dropdown was silently empty in production

[JSONbored]

Summary

User-reported live bug: the AI-usage dashboard's Provider/Feature/Model filters showed red error
icons and couldn't actually filter anything, and several panels (tokens-by-feature, cost-share-by-
feature) showed no data despite the dashboard's own summary panels clearly showing real, non-zero
totals.

Root cause, confirmed empirically against the live frser-sqlite-datasource plugin via
POST /api/ds/query: every "query"-type template variable across every dashboard
($repo, $scope, $provider/$feature/$model, added across #5234/#5267/#5302) had its
query object carry only rawQueryText — a display/round-trip field the plugin's variable-query
path never actually executes — never queryText, the field its query builder actually requires.
Every one of these dropdowns silently resolved to zero options in production; the identical raw SQL
run directly via the sqlite3 CLI, or via the API with queryText added, returns real data
immediately. Panel targets were never affected — they already set both fields at the top level —
which is exactly why every table/timeseries panel kept rendering real data while every variable
dropdown was broken, cascading into "No data"/zeroed panels wherever a panel's own WHERE clause
referenced one of the broken variables.

Since these are file-provisioned dashboards (editable: false), the fix couldn't be hotfixed via the
dashboard-save API — I updated the files directly on edge-nl-01 and restarted Grafana to confirm the
fix live before opening this PR (see test plan).

Adds a cross-dashboard regression test scanning every file in grafana/dashboards/ for this exact
shape, so a future dashboard can't reintroduce it by copying the variable pattern without the fix.

Test plan

• npm run typecheck — clean
• npm run test:ci — full local gate green
• New selfhost-grafana-variable-query-text.test.ts: scans every dashboard file for this exact
  missing-queryText shape; verified it fails when the bug is reintroduced (temporarily reverted
  one file, confirmed the test catches it) and passes with the fix
• Live-verified on edge-nl-01 via POST /api/ds/query against the running
  frser-sqlite-datasource: the $provider variable's query returned zero rows before the fix,
  real provider names after; the previously-empty "cost share by feature" panel query also
  confirmed to return real, non-zero data with the fix in place

[superagent-security]

Superagent didn't find any vulnerabilities or security issues in this PR.

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 94.36%. Comparing base (b779f28) to head (593e698).
✅ All tests successful. No failed tests found.

Additional details and impacted files

@@            Coverage Diff             @@
##             main    #5349      +/-   ##
==========================================
- Coverage   94.40%   94.36%   -0.04%     
==========================================
  Files         550      550              
  Lines       44119    44119              
  Branches    14632    14632              
==========================================
- Hits        41652    41635      -17     
- Misses       1792     1809      +17     
  Partials      675      675              

Flag	Coverage Δ
shard-1	43.75% <ø> (-0.26%)	⬇️
shard-2	34.13% <ø> (-0.52%)	⬇️
shard-3	32.16% <ø> (+0.55%)	⬆️
shard-4	31.21% <ø> (-0.21%)	⬇️
shard-5	33.04% <ø> (-0.12%)	⬇️
shard-6	43.63% <ø> (+0.27%)	⬆️

Flags with carried forward coverage won't be shown. Click here to find out more.
see 1 file with indirect coverage changes

🚀 New features to boost your workflow:

• 📦 JS Bundle Analysis: Save yourself from yourself by tracking and limiting bundle sizes in JS merges.

[gittensory-orb]

Warning

🟨🟨🟨🟨🟨🟨🟨🟨🟨🟨🟨🟨

⏸️ Gittensory review result - manual review recommended

_{Review updated: 2026-07-12 14:45:52 UTC}

4 files · 1 AI reviewer · 2 blockers · readiness 93/100 · CI green · unstable

⏸️ Suggested Action - Manual Review

• No linked issue detected — If this PR is intended to solve an issue, link it explicitly in the PR body.
• Maintainer requires a linked issue — Link the relevant issue (for example Closes #123) before opening the PR.

Review summary
The diff adds `queryText` (mirroring the existing `rawQueryText`) to the query-type template variables in three provisioned dashboards, matching the documented root cause that the frser-sqlite-datasource plugin's variable-query path reads `queryText`, not `rawQueryText`. The fix is applied consistently across all four affected variables ($repo/$scope-equivalents, $provider, $feature, $model) and is backed by a new cross-dashboard regression test that scans every dashboard file for the missing-queryText shape, guarding against recurrence. The existing test file shown in context still only asserts on `rawQueryText`, so it will keep passing without validating the new `queryText` field, but that's a pre-existing test, not something this diff needs to fix.

Blockers

• PR description: you need to close or clearly link the eligible open issue for this user-reported production bug, for example `Closes feat(orb): add POST /v1/orb/ingest central collector endpoint #1234`, or explain why this PR is maintainer-authorized without an issue.

Nits — 5 non-blocking

• The new regression test in test/unit/selfhost-grafana-variable-query-text.test.ts only checks variables whose datasource type is exactly 'frser-sqlite-datasource'; if a dashboard's per-variable datasource block is ever omitted (inheriting the panel/dashboard-level datasource) the check would silently skip that variable — worth a comment noting this is a known scanning limitation.
• The pre-existing selfhost-grafana-ai-usage-dashboard.test.ts (test/unit/selfhost-grafana-ai-usage-dashboard.test.ts) still only asserts on `query?.rawQueryText`; consider updating it to also assert `queryText` so a future accidental removal of `queryText` alone wouldn't slip past that suite.
• github-prs.json's added queryText is a large inlined SQL string duplicated byte-for-byte from rawQueryText — acceptable given the plugin's requirement, but a helper/constant to keep the two in sync would reduce future copy-paste drift risk.
• Consider having the new test also verify queryText against a golden/expected string per known variable name (like the existing ai-usage test does for rawQueryText) rather than only structural equality with rawQueryText, to catch cases where both fields could be updated together but incorrectly.
• Add a one-line PR-description or code comment cross-reference between the two test files so future readers understand why there are two overlapping Grafana dashboard test suites.

Concerns raised — review before merging

• No linked issue detected — If this PR is intended to solve an issue, link it explicitly in the PR body.
• Maintainer requires a linked issue — Link the relevant issue (for example Closes #123) before opening the PR.

Signal	Result	Evidence
Code review	❌ 2 blockers	1 reviewer
Linked issue	⚠️ Missing	No linked issue or no-issue rationale found.
Related work	✅ No active overlap found	No same-issue or scoped active PR overlap found.
Change scope	✅ 20/20	Low review scope from cached public metadata (no linked issue context).
Validation posture	✅ 25/25	PR body includes validation/test evidence.
Contributor workload	✅ 10/10	Author activity: 44 registered-repo PR(s), 36 merged, 476 issue(s).
Contributor context	✅ Confirmed Gittensor contributor	JSONbored; Gittensor profile; 44 PR(s), 476 issue(s).
Gate result	❌ Blocking	Repo-configured hard blocker found.
Improvement	⚠️ ℹ️ Insufficient signal	risk: clean · value: insufficient-signal — Nothing measurable for the structural-improvement analyzers on this PR (e.g. no code files changed).

Review context

• Author: JSONbored
• Role context: owner (maintainer lane)
• Public audience mode: oss maintainer
• Lane context: Repository is configured for direct PR review.
• Public profile languages: not available
• Official Gittensor activity: 44 PR(s), 476 issue(s).
• PR-specific overlap: none found.

Contributor next steps

• Treat this as maintainer-lane context rather than normal contributor-lane activity.
• Explain no-issue PR.
• Link the issue being solved, or explicitly explain why this is a no-issue PR.

Signal definitions

• Related work = same linked issue, overlapping active PRs, or title/path similarity.
• Change scope = cached public metadata such as size labels, draft state, and review-burden hints.
• Validation posture = whether the PR provides enough public validation/test evidence for maintainer review.
• Contributor workload = public contributor activity and cleanup pressure, not a repo-wide quality failure.
• Contributor context = public GitHub/Gittensor identity context; non-Gittensor status is not a blocker.

[BETA] Chat with Gittensory

Ask Gittensory a question about this PR directly in a comment — grounded only in the same cached, public-safe facts shown above, never a new claim.

• @gittensory ask <question> answers contribution-quality Q&A with source citations and freshness.
• @gittensory chat <question> answers in natural prose from cached decision-pack facts via local inference (maintainer/collaborator; read-only).
• A plain-language @gittensory mention with a real question is routed to the closest matching read-only command automatically -- no exact syntax required.

Full command reference: https://gittensory.aethereal.dev/docs/gittensory-commands

_{🟩 Safe / merged · 🟦 Advisory · 🟨 Held for review · 🟥 Blocked / closed}

---

💰 Earn for open-source contributions like this. Gittensor lets GitHub contributors earn for the work they already do — register to start earning →.

Checked by Gittensory, a quiet PR intelligence layer for OSS maintainers.

• Re-run Gittensory review