feat(setbuilder): TrackVibe LLM enrichment + community vibe display (read) (#391)#422
Merged
Conversation
… note - community_vibe.py: module docstring notes that override rows are FULL SNAPSHOTS and the v1.1 write endpoint must carry forward unchanged fields; CommunityVibe dataclass docstring defines sample_size. - vibe_resolver.py: dedupe keys via dict.fromkeys before IN-clause queries in _own_overrides and _llm_vibes; community_consensus already deduped in-place. - community_vibe.py: community_consensus dedupes track_keys via dict.fromkeys so tracks sharing a vibe_key don't generate duplicate IN values. - test_setbuilder_vibes.py: three edge-case tests added to TestCommunityConsensus: stddev exactly at threshold is rejected, single-vote with min_sample=1 passes, all-null retraction row supersedes the earlier vote (retraction test).
GET /api/setbuilder/sets/{id}/pool/vibes returns the three-tier vibe
state (own / community / LLM, plus per-field resolution with source
provenance) for every pool track. POST .../pool/vibes/enrich runs the
batch LLM enrichment through the gateway (5/min — each call fans out
to BATCH_SIZE-track dispatches) and returns run stats plus the
refreshed state. NoLlmConfigured maps to a friendly 400. Both
endpoints keep owner-private 404 semantics.
Migration 057 (1899c49) added vibe_consensus_min_sample / vibe_consensus_max_stddev to SystemSettings but the exact-key contract test was not updated, failing on branch since that commit.
|
Caution Review failedThe pull request is closed. ℹ️ Recent review info⚙️ Run configurationConfiguration used: defaults Review profile: CHILL Plan: Pro Plus Run ID: 📒 Files selected for processing (1)
📝 WalkthroughWalkthroughAdds three‑tier TrackVibe support: DB migration and settings, backend consensus/enrichment/resolver services, new setbuilder GET/POST pool-vibes API and OpenAPI/TS surface, VibeTiers UI + CSS, PoolPanel toggle/analyze integration, and extensive tests and docs. ChangesTrackVibe Enrichment Feature
Sequence Diagram(s)sequenceDiagram
participant User
participant PoolPanel
participant API_Get as GET /pool/vibes
participant API_Enrich as POST /pool/vibes/enrich
participant Resolver
participant DB
participant Gateway
User->>PoolPanel: click "Vibes"
PoolPanel->>PoolPanel: toggleVibes()
alt vibes not loaded
PoolPanel->>API_Get: request PoolVibesState
API_Get->>Resolver: build_pool_vibe_states(set_id)
Resolver->>DB: read overrides, community votes, LLM cache
Resolver-->>API_Get: PoolVibesState
API_Get-->>PoolPanel: PoolVibesState
PoolPanel->>PoolPanel: buildVibeMap()
end
User->>PoolPanel: click "Analyze"
PoolPanel->>API_Enrich: request enrichment
API_Enrich->>DB: check cache, determine missing keys
API_Enrich->>Gateway: dispatch batched tool calls for missing tracks
Gateway-->>API_Enrich: tool-call responses
API_Enrich->>DB: persist TrackVibe rows
API_Enrich-->>PoolPanel: VibeEnrichmentResult + refreshed PoolVibesState
PoolPanel->>PoolPanel: update vibes map, show toast
Estimated code review effort🎯 4 (Complex) | ⏱️ ~60 minutes Possibly related PRs
🚥 Pre-merge checks | ✅ 4 | ❌ 1❌ Failed checks (1 warning)
✅ Passed checks (4 passed)
✏️ Tip: You can configure your own custom pre-merge checks in the settings. ✨ Finishing Touches📝 Generate docstrings
🧪 Generate unit tests (beta)
Comment |
Collaborator
Author
|
@coderabbitai review |
✅ Action performedReview finished.
|
Collaborator
Author
|
@coderabbitai full review |
✅ Action performedFull review finished. |
![coderabbitai[bot]](https://avatars.githubusercontent.com/in/347564?s=60&v=4){kind=small}
There was a problem hiding this comment.
Actionable comments posted: 3
🧹 Nitpick comments (2)
server/tests/test_setbuilder_vibes_api.py (1)
59-59: 💤 Low valueConsider adding return type annotation for consistency.
The helper
_llm_vibe_rowis missing a return type annotation, while_fake_responseon line 72 includes one. Adding-> TrackVibewould improve consistency and readability.♻️ Suggested fix
-def _llm_vibe_row(track_id: str = "tidal:0", *, energy: int = 5, confidence: float = 0.8): +def _llm_vibe_row(track_id: str = "tidal:0", *, energy: int = 5, confidence: float = 0.8) -> TrackVibe:🤖 Prompt for AI Agents
Verify each finding against current code. Fix only still-valid issues, skip the rest with a brief reason, keep changes minimal, and validate. In `@server/tests/test_setbuilder_vibes_api.py` at line 59, The helper function _llm_vibe_row lacks a return type annotation—add "-> TrackVibe" to its signature to match the style of _fake_response and improve consistency; update the function definition for _llm_vibe_row(track_id: str = "tidal:0", *, energy: int = 5, confidence: float = 0.8) to include the return type TrackVibe and ensure any imports or type references needed for TrackVibe are available in the test file.dashboard/app/(dj)/setbuilder/setbuilder.module.css (1)
619-619: ⚡ Quick winUse
remunits for font-size to match file conventions and improve accessibility.Line 619 declares
font-size: 10px, but every other font-size in this file usesremunits (e.g., line 359 uses0.625remfor the same 10px equivalent). Usingremrespects user browser font-size preferences and maintains consistency with the codebase.♻️ Proposed fix
.vibeChip { display: inline-flex; align-items: center; gap: 4px; - font-size: 10px; + font-size: 0.625rem; padding: 1px 6px; border-radius: 4px; border: 1px solid var(--border, `#3a3a3a`); color: var(--text-secondary); }🤖 Prompt for AI Agents
Verify each finding against current code. Fix only still-valid issues, skip the rest with a brief reason, keep changes minimal, and validate. In `@dashboard/app/`(dj)/setbuilder/setbuilder.module.css at line 619, Replace the hard-coded "font-size: 10px" in the affected rule in setbuilder.module.css with the equivalent rem unit used throughout the file (0.625rem) so it matches the project's rem-based convention and respects user/browser font-size settings.
🤖 Prompt for all review comments with AI agents
Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.
Inline comments:
In `@dashboard/app/`(dj)/setbuilder/components/PoolPanel.tsx:
- Around line 163-176: The toggleVibes handler sets vibesLoaded before the async
getPoolVibes completes, so a failed fetch leaves vibesLoaded true and prevents
retries; change the flow in toggleVibes (and related state calls) so you do not
set vibesLoaded(true) up-front — instead call setVibesBusy(true) then kick off
api.getPoolVibes(setId) and on success call setVibes(buildVibeMap(...)) and
setVibesLoaded(true); on failure clear vibesBusy and also ensure vibesLoaded is
false (or not set) and propagate the toast; keep the final finally to clear
vibesBusy. Reference: toggleVibes, api.getPoolVibes, setVibesLoaded,
setVibesBusy, setVibes, setToast.
In `@dashboard/lib/api-types.generated.ts`:
- Around line 11217-11247: The generated operation
enrich_pool_vibes_api_setbuilder_sets__set_id__pool_vibes_enrich_post is missing
the documented 400 response for NoLlmConfigured; update that operation's
responses to include a 400 entry (headers and content "application/json")
pointing to the appropriate error schema (e.g.,
components["schemas"]["HTTPError"] or the specific NoLlmConfigured schema) so
the frontend can type-check and handle the user-facing 400 error.
In `@server/openapi.json`:
- Around line 16492-16512: Add the missing 400 response for POST
/pool/vibes/enrich by updating the route definition to include a 400 response
that references the friendly error schema (e.g., "NoLlmConfigured") and then
regenerate the OpenAPI spec so server/openapi.json includes that response;
specifically, modify the POST /pool/vibes/enrich route source to add a
responses["400"] entry with description "No LLM configured" and content
application/json -> schema -> $ref to the NoLlmConfigured (or equivalent
friendly error) component, then run the OpenAPI generation step to produce an
updated openapi.json containing the 400 entry.
---
Nitpick comments:
In `@dashboard/app/`(dj)/setbuilder/setbuilder.module.css:
- Line 619: Replace the hard-coded "font-size: 10px" in the affected rule in
setbuilder.module.css with the equivalent rem unit used throughout the file
(0.625rem) so it matches the project's rem-based convention and respects
user/browser font-size settings.
In `@server/tests/test_setbuilder_vibes_api.py`:
- Line 59: The helper function _llm_vibe_row lacks a return type annotation—add
"-> TrackVibe" to its signature to match the style of _fake_response and improve
consistency; update the function definition for _llm_vibe_row(track_id: str =
"tidal:0", *, energy: int = 5, confidence: float = 0.8) to include the return
type TrackVibe and ensure any imports or type references needed for TrackVibe
are available in the test file.
🪄 Autofix (Beta)
Fix all unresolved CodeRabbit comments on this PR:
- Push a commit to this branch (recommended)
- Create a new PR with the fixes
ℹ️ Review info
⚙️ Run configuration
Configuration used: defaults
Review profile: CHILL
Plan: Pro Plus
Run ID: e53711ae-f2b8-49b1-bf86-9afd0d87139e
📒 Files selected for processing (27)
dashboard/app/(dj)/setbuilder/components/PoolPanel.tsxdashboard/app/(dj)/setbuilder/components/VibeTiers.tsxdashboard/app/(dj)/setbuilder/components/__tests__/PoolPanel.test.tsxdashboard/app/(dj)/setbuilder/components/__tests__/VibeTiers.test.tsxdashboard/app/(dj)/setbuilder/setbuilder.module.cssdashboard/app/admin/settings/__tests__/page.test.tsxdashboard/lib/api-types.generated.tsdashboard/lib/api-types.tsdashboard/lib/api.tsdocs/superpowers/plans/2026-06-10-trackvibe-llm-enrichment.mdserver/alembic/versions/057_add_vibe_consensus_settings.pyserver/app/api/admin.pyserver/app/api/setbuilder.pyserver/app/models/system_settings.pyserver/app/schemas/setbuilder.pyserver/app/schemas/system_settings.pyserver/app/services/llm/base.pyserver/app/services/llm/gateway.pyserver/app/services/setbuilder/community_vibe.pyserver/app/services/setbuilder/vibe_enrichment.pyserver/app/services/setbuilder/vibe_resolver.pyserver/app/services/system_settings.pyserver/openapi.jsonserver/tests/test_api_contracts.pyserver/tests/test_llm_gateway.pyserver/tests/test_setbuilder_vibes.pyserver/tests/test_setbuilder_vibes_api.py
…h endpoint - PoolPanel: only set vibesLoaded after a successful getPoolVibes, so a failed initial fetch can be retried on the next toggle-on; guard with vibesBusy to prevent duplicate in-flight GETs. Regression test added. - Declare the 400 (NoLlmConfigured) response on POST /pool/vibes/enrich and regenerate openapi.json + api-types.generated.ts so the typed contract includes the error case the frontend special-cases. Addresses CodeRabbit review on PR #422.
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Closes #391
Summary
services/setbuilder/vibe_enrichment.py): enriches uncached pool tracks via the LLM gateway with forcedtool_use(20 tracks/call,purpose="vibe_enrichment"), writing to the globalTrackVibecache (table pre-existed from the Phase-0 scaffold). 100 tracks cost exactly 5 gateway calls; a second DJ with the same tracks pays zero (test-pinned).services/setbuilder/community_vibe.py): aggregatesTrackVibeOverridevotes (latest row per track+user); energy consensus gated onsample_size >= 3 AND pstdev < 1.5, mood on strict majority — both thresholds admin-tunable via newSystemSettingscolumns (migration 057,down_revision="056").services/setbuilder/vibe_resolver.py): per-field cascade own override → community consensus → LLM cache, withlow_confidenceflagging (confidenceNone or < 0.5).GET /api/setbuilder/sets/{id}/pool/vibes(60/min) andPOST .../pool/vibes/enrich(5/min,NoLlmConfigured→ friendly 400). Owner-private 404 semantics. OpenAPI + generated dashboard types regenerated.Vibestoggle +Analyzebutton; read-onlyVibeTierschips (You / Crowd / AI) per track, sample-size on the community chip, per-field winner highlight, low-confidence AI guesses flagged with dashed-amber ⚠ styling.ChatResponse.providerpopulated from the resolved connector'sconnector_typeso the cache can storellm_provider/llm_modelprovenance.Design decisions
model_hint="fast"deviation:ChatRequesthas no speed-tier concept — dispatch usesmodel=None, so the DJ's connector-configured model governs. A cross-provider fast/smart mapping would mean refactoring the gateway (out of scope); documented in-code.ChatResponsedidn't surface which connector served a call. Minimal additiveproviderfield, populated in the gateway's_attemptsuccess path viamodel_copy— no other gateway behavior changed (streaming path intentionally untouched).SetPoolTrack.track_idis nullable, so the fallback key issig:{dedupe_sig}(normalized artist+title hash) — globally stable, consistent with the namespaced free-formtrack_idconvention, and lets manual tracks share the global cache.TrackViberow matching(track_id, prompt_version, schema_version)counts regardless of provider/model — that's what makes "second DJ pays zero" true across different connectors. Newest row wins at read time; bumpingPROMPT_VERSIONin code lazily invalidates old rows.LlmErrormarks the current batch + all remaining tracks failed and stops (no provider hammering);NoLlmConfiguredmaps to a friendly 400. A mid-batch insert race keeps the paid LLM rows (re-query winners, re-insert only missing — test-pinned).TrackVibeOverrideexisted from the Phase-0 scaffold (migration 046), so no new table was needed; per the issue note, no write endpoints were added (write UX is v1.1) — this PR is the read path only.LlmVibeOut.low_confidence) so the frontend stays dumb; threshold 0.5.PATCH /api/admin/settings(bounds 1–100 / 0.1–5.0, 422-tested); admin-UI form controls deferred.Test plan
alembic upgrade head && alembic check(single head 057), pytest — 2749 passed, coverage 88.38% (≥85%)tsc --noEmit, vitest — 1118 passed, coverage thresholds holdtest_100_tracks_costs_5_calls,test_second_run_fully_cached_even_for_other_dj, resolver tier tests (test_own_override_wins/test_community_beats_llm/test_llm_fallback/test_no_tiers_resolves_none)openapi.json+api-types.generated.tsverified byte-identical to regenerationSummary by CodeRabbit
New Features
Chores
Tests
Docs