feat(commands): make @gittensory chat's frontier fallback configurable (#4595 follow-up)

[JSONbored]

Summary

Follow-up to #4595 (@gittensory chat). That PR shipped chatQa as strictly Ollama-only, with no fallback path — this adds a companion opt-in flag for self-hosters who'd rather use their own frontier subscription/tokens for chat than run a local GPU.

• New advisoryAiRouting.chatQaFrontierFallback: boolean (default false). When env.AI_ADVISORY is unconfigured and this is explicitly enabled, chat falls back to the shared frontier env.AI chain instead of declining.
• Ollama stays strictly preferred: even with the flag on, env.AI_ADVISORY is always tried first — the frontier chain is only ever touched when Ollama is genuinely unavailable.
• Global default + per-repo override, identical shape to every other advisoryAiRouting capability (yml-only, config-as-code, both normalizers + engine-package twin + focus-manifest merge + OpenAPI schema updated).
• Usage/audit events now record which provider actually served a given answer (provider: "advisory" | "frontier"), so an operator who enables this can see at a glance whether they're spending frontier tokens.

Also verified and closed a real gap while implementing this: chatQa had never actually been activated on any production repo (the capability existed in code from #4595 but no .gittensory.yml set it), and no repo had commandRateLimitPolicy turned on — meaning the command would have had zero real rate-limit ceiling if enabled. Both are now fixed on the self-host deployment (this repo + awesome-claude): chatQa: true, chatQaFrontierFallback left unset (Ollama-only, matching this deployment's own preference), commandRateLimitPolicy: hold so the existing 5-per-24h AI_COST_BEARING_COMMANDS ceiling actually applies.

Test plan

• New tests: frontier-fallback success path, Ollama-still-preferred-when-fallback-enabled, neither-provider-configured message, plus the untouched default-off behavior re-asserted
• Capability-flag normalization (it.each field lists) + focus-manifest sparse-parse/merge tests, both app and engine-package copies
• npm run typecheck, npm run test:coverage (unsharded, 100% on every changed line/branch), npm run test:ci, npm audit --audit-level=moderate all green
• npm run ui:openapi regenerated and committed
• npx tsx scripts/check-engine-parity.ts passes

[superagent-security]

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

[cloudflare-workers-and-pages]

Deploying with    Cloudflare Workers

The latest updates on your project. Learn more about integrating Git with Workers.

Status	Name	Latest Commit	Preview URL	Updated (UTC)
✅ Deployment successful! View logs	gittensory-ui	362c203	Commit Preview URL Branch Preview URL	Jul 11 2026, 08:58 AM

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 94.34%. Comparing base (a086033) to head (362c203).
⚠️ Report is 2 commits behind head on main.
✅ All tests successful. No failed tests found.

Additional details and impacted files

@@           Coverage Diff           @@
##             main    #5015   +/-   ##
=======================================
  Coverage   94.34%   94.34%           
=======================================
  Files         456      456           
  Lines       39188    39196    +8     
  Branches    14290    14297    +7     
=======================================
+ Hits        36970    36978    +8     
  Misses       1570     1570           
  Partials      648      648           

Flag	Coverage Δ
shard-1	46.81% <0.00%> (-0.01%)	⬇️
shard-2	33.42% <15.38%> (+0.15%)	⬆️
shard-3	31.49% <0.00%> (+0.22%)	⬆️
shard-4	32.74% <84.61%> (-0.40%)	⬇️
shard-5	33.56% <0.00%> (-0.01%)	⬇️
shard-6	45.16% <15.38%> (-0.01%)	⬇️

Flags with carried forward coverage won't be shown. Click here to find out more.

Files with missing lines	Coverage Δ
packages/gittensory-engine/src/focus-manifest.ts	99.17% <100.00%> (+<0.01%)	⬆️
...ry-engine/src/review/advisory-ai-routing-config.ts	100.00% <ø> (ø)
src/openapi/schemas.ts	100.00% <ø> (ø)
src/review/advisory-ai-routing-config.ts	100.00% <ø> (ø)
src/services/ai-chat-qa.ts	100.00% <100.00%> (ø)
src/signals/focus-manifest.ts	99.64% <100.00%> (+<0.01%)	⬆️
src/types.ts	100.00% <ø> (ø)

🚀 New features to boost your workflow:

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

[loopover-orb]

Warning

🟨🟨🟨🟨🟨🟨🟨🟨🟨🟨🟨🟨

⏸️ Gittensory review result - manual review recommended

_{Review updated: 2026-07-11 09:04:38 UTC}

16 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
This PR adds an opt-in `advisoryAiRouting.chatQaFrontierFallback` flag so `@​gittensory chat` can fall back to the shared frontier `env.AI` when `env.AI_ADVISORY` is unconfigured, while still preferring Ollama first. The core logic in `src/services/ai-chat-qa.ts` (`env.AI_ADVISORY ?? (frontierFallbackAllowed ? env.AI : undefined)`) correctly preserves Ollama-first behavior and only touches the frontier chain when explicitly opted in, and the flag is wired consistently through every layer this repo requires for a new `advisoryAiRouting` capability: type twins (`src/types.ts` and `packages/gittensory-engine/src/types/manifest-deps-types.ts`), both normalizer copies, the zod schema, OpenAPI spec, focus-manifest sparse-override parsing, and `resolveEffectiveSettings` merge — matching the identical pattern used for `chatQa` itself. Tests cover both branches (Ollama-preferred-even-when-fallback-enabled, frontier-used-when-Ollama-absent, neither-configured) plus the default-off regression case, so the new logic is real coverage, not fabricated.

Nits — 5 non-blocking

• The PR description claims `.gittensory.yml` changes for this repo and awesome-claude (`chatQa: true`, `commandRateLimitPolicy: hold`) were made, but no such file diff is visible in the provided changeset — only the `.gittensory.yml.example`/`config/examples/gittensory.full.yml` docs are touched; confirm those live config changes actually shipped in this PR or a companion commit.
• The description doesn't link an eligible open issue for this follow-up work, only referencing PR feat(commands): add @gittensory chat <question> — grounded LLM Q&A via local Ollama #4595 — worth a maintainer note per repo convention even though the change itself is narrow and well-scoped.
• The `advisory-ai-routing-config.ts` normalizer is duplicated verbatim between `src/review/` and `packages/gittensory-engine/src/review/` (both edited identically here); pre-existing twin-file pattern, not introduced by this diff, but a DRY candidate.
• Consider a short comment or CHANGELOG note pointing at wherever `chatQa`/`chatQaFrontierFallback` get enabled in this repo's actual `.gittensory.yml`, so the audit trail matches the description's claim.
• If the twin normalizer files (`src/review/advisory-ai-routing-config.ts` vs `packages/gittensory-engine/src/review/advisory-ai-routing-config.ts`) are meant to converge, this would be a reasonable moment to extract a shared implementation given how mechanically identical the two diffs are.

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: 47 registered-repo PR(s), 39 merged, 428 issue(s).
Contributor context	✅ Confirmed Gittensor contributor	JSONbored; Gittensor profile; 47 PR(s), 428 issue(s).
Gate result	❌ Blocking	Repo-configured hard blocker found.
Improvement	✅ Minor	risk: clean · value: minor — Code changes are accompanied by test evidence.

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: 47 PR(s), 428 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.

Visual preview

Route	Viewport	Before (production)	After (this PR's preview)	Diff
/	desktop			—
/	mobile			—

Click any thumbnail to open the full-size screenshot. Before = production · After = this PR's preview deploy.

_{🟩 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