pjandro/OmniRoute
1
0
Fork 0
mirror of https://github.com/diegosouzapw/OmniRoute.git synced 2026-08-03 13:52:09 +03:00
Code Issues Packages Projects Releases Wiki Activity
4,934 Commits 110 Branches 313 Tags
2d321e1f52ec00d16ad016edf59cd406cd031f98
Commit Graph

4934 Commits

This Branch
This Branch
All Branches
Author SHA1 Message Date
diegosouzapw
3cb3b88c9d merge(F9): docs (AGENTBRIDGE/TRAFFIC_INSPECTOR) + openapi + E2E specs + CHANGELOG (Group A) 2026-05-28 10:26:02 -03:00
diegosouzapw
80910714a3 Merge branch 'feat/playground-ui-core-F6' into feat/playground-ui-advanced-F7 2026-05-28 10:07:46 -03:00
diegosouzapw
24e251fff2 Merge branch 'feat/playground-hooks-F5' into feat/playground-ui-advanced-F7 2026-05-28 10:07:36 -03:00
diegosouzapw
d2a0097f86 Merge branch 'feat/playground-search-foundation-F1' into feat/playground-ui-advanced-F7 2026-05-28 10:07:30 -03:00
diegosouzapw
6f7b051289 merge(F8): MCP strategy-from-settings + CLI canonical types 2026-05-28 10:07:04 -03:00
diegosouzapw
edaa05783d fix(memory): correct operator precedence in listEmbeddingProviders hasKey check (TS18047) 
The previous expression was parsed as (A && B && C) || D, allowing D to evaluate
with creds possibly null. Wrap (apiKey || accessToken) in parens so creds-narrowing
covers the whole disjunction.
 2026-05-28 10:06:47 -03:00
oyi77
997ece8ef0 feat(gitlawb): dynamic model fetch with gmi-cloud fallback 
Hybrid approach:
- gitlawb (xiaomi-mimo): dynamic /models endpoint → 356 models
- gitlawb-gmi (gmi-cloud): 404 fallback → local catalog gracefully
Mimics Gitlawb/openclaude's model-routing pattern
 2026-05-28 20:05:11 +07:00
diegosouzapw
4640f0a77b feat(dashboard): refactor search-tools into Studio UI with 3 tabs (F8) 
- SearchToolsClient: Studio orchestrator with Search/Scrape/Compare tabs
  and shared SearchToolsConfigPane; latency+cost metrics wired to TopBar
- SearchToolsTopBar: 3-tab switcher with metrics display and Export Code
  button (MockExportCodeModal placeholder annotated TODO(F7-merge))
- SearchToolsConfigPane: provider catalog inline with per-tab options
  (search-type, fetch format/full-page, rerank model, history)
- SearchConceptCard: collapsible explanatory cards for all 5 modalities
- ProviderCatalog: fetches /api/search/providers, renders 12 search + 3
  fetch providers with status badges (configured/missing/rate_limited)
- ScrapeResult: markdown preview + raw toggle + D21 256KB cap with
  truncation warning and full-content raw modal
- tabs/SearchTab: SearchForm + ResultsPanel + RerankPanel with
  noProvidersConfigured CTA wired
- tabs/ScrapeTab: URL input + /v1/web/fetch call via useScrapeFetch hook
- tabs/CompareTab: up to 4 providers (D22) in parallel via Promise.allSettled,
  overlap calculation, best/worst coloring
- hooks/useScrapeFetch: fetch wrapper for /v1/web/fetch with latency
- SearchForm: extended with catalog provider metadata badges
- ResultsPanel: extended with noProvidersConfigured CTA empty state
- ProviderComparison: "Size" hardcoded text annotated with data-i18n attr
- 7 vitest UI test files: 75 tests covering all acceptance criteria
 2026-05-28 09:44:09 -03:00
diegosouzapw
44cddf8c4a Merge FIX-1 (gap fixes G2-G7: validating spinner, ExpirationBadge in files, polling spinner, i18n confirm, error keys, retention bullet) 2026-05-28 09:40:04 -03:00
diegosouzapw
bc0301503c fix(batch): gap fixes G2-G7 — validating spinner, ExpirationBadge in files, polling spinner, i18n confirm, error keys, retention bullet 2026-05-28 09:39:01 -03:00
diegosouzapw
1974628190 fix(mcp,cli): read retrieval strategy from settings + update CLI memory types (plan 21 F8) 
- memoryTools.ts: replace hardcoded retrievalStrategy:"exact" with getMemorySettings()+toMemoryRetrievalConfig(); fallback to "exact" on catch
- memory.mjs: VALID_TYPES updated to ["factual","episodic","procedural","semantic"]; default changed from "user" to "factual"; legacy types (user/feedback/project/reference) emit deprecation warning and map to "factual"
- tests: mcp-memory-tools-strategy.test.ts (7 cases) + cli-memory-types.test.mjs (12 cases)
 2026-05-28 09:35:52 -03:00
diegosouzapw
5e8c17d0f5 test(playground): UI coverage for studio + tabs 
5 vitest test files covering PlaygroundStudio (smoke, 4 tabs, deep-link),
StudioConfigPane (collapse/expand, sliders, endpoint select), ChatTab (SSE
send + markdown + system prompt propagation + regenerate + onMetricsUpdate),
ApiTab (smoke, 10 endpoints, SSE stream), and TokenCostCounter (all display
states). 42 tests, all passing.
 2026-05-28 09:25:20 -03:00
diegosouzapw
bcfb8968bb refactor(playground): remove ChatPlayground/SearchPlayground (migrated) 
ChatPlayground.tsx → ChatTab (markdown + system prompt + metrics).
SearchPlayground.tsx → superseded by /dashboard/search-tools Studio (F8).
grep confirmed zero external imports before deletion.
 2026-05-28 09:25:09 -03:00
diegosouzapw
9a36d9ecf5 feat(playground): migrate Monaco editor to ApiTab (D14 zero regression) 
Cuts the full content of page.tsx (889 LOC) to tabs/ApiTab.tsx with zero
logic changes: 10 endpoints, multimodal upload (vision images + audio),
SSE streaming, model/provider select, Monaco request/response editors,
image generation inline render, speech playback. page.tsx replaced with
Suspense-wrapped <PlaygroundStudio /> shell.
 2026-05-28 09:24:55 -03:00
diegosouzapw
f24406dcf4 feat(playground): migrate ChatPlayground to ChatTab with markdown + metrics 
Refactors ChatPlayground.tsx into ChatTab — multi-turn SSE chat with:
markdown rendering via MarkdownMessage (F1), system prompt from config pane,
token/cost per message via useStreamMetrics (F5), regenerate button, and
stop/cancel support. Hard Rule compliance: no useCallback to satisfy
react-hooks/preserve-manual-memoization rule.
 2026-05-28 09:24:44 -03:00
diegosouzapw
e2a22c57b0 feat(playground): add ParamSliders + TokenCostCounter 
ParamSliders: temperature/max_tokens/top_p/penalties/seed/stop/JSON-mode
sliders and inputs. TokenCostCounter: D13-compliant "(estimated)" cost label
with ↑/↓ arrow notation (e.g. 142↑ 38↓ · $0.002 estimated).
 2026-05-28 09:24:34 -03:00
diegosouzapw
68ec69a9c5 feat(playground): scaffold PlaygroundStudio + StudioTopBar + StudioConfigPane 
Adds the PlaygroundStudio orchestrator (tab routing via ?tab= deep-link,
shared configState), StudioTopBar (4 tabs + export placeholder modal), and
StudioConfigPane (collapsible, 10-endpoint select, model, system prompt,
ParamSliders). F7 slots (SLOT_PRESETS / SLOT_IMPROVE) left as JSX comments.
 2026-05-28 09:24:21 -03:00
Nikolay Alafuzov
b209387d9f fix(reasoning): guard interleaved capability lookup 2026-05-28 15:23:58 +03:00
oyi77
471df45bcb refactor(gitlawb): extract shared opengateway validator factory, fix docs path in test 
- Extract gitlawb/gitlawb-gmi validators into buildOpengatewayValidator factory
- Fix dockerignore-docs-coverage test: update stale docs/AUTO-COMBO.md -> docs/routing/AUTO-COMBO.md
 2026-05-28 19:21:08 +07:00
diegosouzapw
8926c8bf98 chore(changelog): document Group A AgentBridge + Traffic Inspector (F9) 2026-05-28 09:14:40 -03:00
diegosouzapw
48c31c4ce5 test(e2e): smoke flows for agent-bridge + traffic-inspector + cross (F9) 2026-05-28 09:14:27 -03:00
diegosouzapw
5ff220b655 docs(api): add ~28 routes for agent-bridge + traffic-inspector to openapi.yaml (F9) 2026-05-28 09:14:18 -03:00
diegosouzapw
7a33af8ef7 docs(architecture): register agent-bridge + traffic-inspector in REPOSITORY_MAP (F9) 2026-05-28 09:14:12 -03:00
diegosouzapw
76fa6688ae docs(frameworks): add TRAFFIC_INSPECTOR.md (F9) 2026-05-28 09:14:04 -03:00
diegosouzapw
12cc1bb79c docs(frameworks): add AGENTBRIDGE.md (F9) 2026-05-28 09:13:51 -03:00
diegosouzapw
70c9bf279a fix(F3): pass sourceModel to agentBridgeHook.recordRequestStart 
hookBufferStart was calling recordRequestStart without sourceModel,
causing the field to be null even when extractSourceModel returned
a value from the body. Now forwards the extracted model so the
Traffic Inspector buffer entry is populated correctly.
 2026-05-28 09:06:16 -03:00
oyi77
bbf8d4ccb9 feat(gitlawb): serve models from static registry without API-unavailable warning 
GitLawB's OpenGateway API does not expose a /models endpoint per
provider-path. Previously the models route fell through to the generic
fallback which returned static catalog models with the misleading
'API unavailable — using local catalog' warning.

Now gitlawb and gitlawb-gmi are handled as static model providers
(same pattern as reka and qwen OAuth) — models are served from the
provider registry without any warning, since all registered models
are functional via POST /chat/completions.
 2026-05-28 18:48:05 +07:00
oyi77
ecb0d1dbac test(gitlawb): add 12 unit tests for gitlawb and gitlawb-gmi specialty validators 
Covers success, auth failure (401/403), non-auth acceptance (400/422/429),
network errors, and custom baseUrl overrides for both providers.
 2026-05-28 18:44:37 +07:00
diegosouzapw
b414df62d5 merge: F9 UI quota-share refactor (DB-backed + Plans page + sidebar + i18n) 2026-05-28 08:29:31 -03:00
diegosouzapw
f78ede0324 merge(F5): memory core rewire (retrieval+store+settings+summarization+reindex) 2026-05-28 08:29:10 -03:00
diegosouzapw
8f6651d053 feat(memory): rewire retrieval/store/settings/summarization + reindex (plan 21 F5) 
- retrieval.ts — semantic/hybrid usa vectorStore quando disponível; degrada para FTS5 transparente
- retrieval.ts — adiciona retrievePreview() (dry-run para Playground) e engineStatus()
- retrieval.ts — rerank opcional via provider configurado (D13)
- store.ts — createMemory/updateMemory geram vetor best-effort; deleteMemory sincroniza vec + Qdrant (D15)
- settings.ts — 7 campos novos (embeddingSource, embeddingProviderModel, transformersEnabled, staticEnabled, rerankEnabled, rerankProviderModel, vectorStore) com defaults
- summarization.ts — summarizeMemoriesOlderThan exposta para uso manual (D19)
- reindex.ts (novo) — runReindexBatch processa fila lazy de backfill (D21)
- 9 testes unitários adicionados; testes F1-F4 sem regressão
 2026-05-28 08:27:28 -03:00
diegosouzapw
a4ee78322f test(ui): cover quota-share page, components, hooks, plans config + sidebar (B/F9) 2026-05-28 08:25:05 -03:00
diegosouzapw
c21bfd5a36 chore(i18n): extend quotaShare + add quotaPlans namespace (pt-BR + en) (B/F9) 2026-05-28 08:25:00 -03:00
diegosouzapw
4cb50a733e feat(sidebar): add costs-quota-plans item to Costs section (B/F9) 2026-05-28 08:24:54 -03:00
diegosouzapw
bd2cf82e0a feat(quota-plans): add /dashboard/costs/quota-share/plans page (B/F9) 2026-05-28 08:24:49 -03:00
diegosouzapw
c7ee32186c refactor(quota-share): move pools from localStorage to /api/quota/pools (B/F9) 2026-05-28 08:24:44 -03:00
diegosouzapw
ec876883b2 feat(quota-share): add hooks (usePools + usePoolUsage + useLocalStoragePoolMigration) (B/F9) 2026-05-28 08:24:38 -03:00
diegosouzapw
160f0693f3 feat(quota-share): extract PoolCard + DimensionBar + AllocationTable + BurnRateChart + ConceptCard + Modals (B/F9) 2026-05-28 08:24:32 -03:00
diegosouzapw
6f62311778 Merge F10 into parent: audit report (GREEN) + docs + micro-fixes for stale cli-tools paths (plan 14) 2026-05-28 08:22:01 -03:00
diegosouzapw
bb90dda1ca chore(plan-14): audit report + docs + plan11 MITM backlog cross-ref (plan 14 F10) 
- docs/reference/CLI-TOOLS.md: update to v3.8.6 (3 pages, unified catalog, MITM backlog, batch API)
- src/shared/components/cli/CliToolCard.tsx: fix stale import path cli-tools → cli-code (audit micro-fix)
- tests/unit/ui/CliToolCard.test.tsx: align mock path to match production import (audit micro-fix)
- src/app/(dashboard)/dashboard/providers/[id]/page.tsx: update stale link /dashboard/cli-tools → /cli-code
- .source/browser.ts + .source/server.ts: auto-regenerated fumadocs MDX index (CLI-TOOLS.md update)
- _tasks/features-v3.8.6/refactorpages/_orchestration/audit-report-14.md: comprehensive audit (gitignored)
- _tasks/features-v3.8.6/refactorpages/_orchestration/_plan11-mitm-backlog.md: updated date/format (gitignored)

Audit conclusion: GREEN — 0 Hard Rule violations, 217+98 tests passing, coverage 56.7/66.2/50.6/56.7 (gate 40/40/40/40).
 2026-05-28 08:19:26 -03:00
diegosouzapw
039ff0abd9 merge(F8): Traffic Inspector UI into Group A parent 2026-05-28 08:05:58 -03:00
oyi77
017c16c80a fix(gitlawb): add specialty validators for connection test — bypass /models probe 
GitLawB OpenGateway API (xiaomi-mimo compatible) does not expose a /models
endpoint, causing validateOpenAILikeProvider to 404 on the initial probe
and report 'Provider validation endpoint not supported'.

Add specialty validators for both gitlawb and gitlawb-gmi that follow the
same pattern as the existing xiaomi-mimo validator: skip GET /models,
validate directly via POST /chat/completions with a minimal test message.
Any 401/403 response means an invalid key; all other responses mean auth
is OK.

Fixes test-connection returning 404 for GitLawB providers.
 2026-05-28 17:57:13 +07:00
diegosouzapw
07d40ef035 merge: F8 REST quota routes (/api/quota/** + /api/settings/quota-store) 2026-05-28 07:55:00 -03:00
diegosouzapw
37e6570bdb test(integration): cover quota REST routes + error sanitization (B/F8) 2026-05-28 07:50:03 -03:00
diegosouzapw
0107beb86b feat(api): add /api/settings/quota-store driver settings route (B/F8) 2026-05-28 07:49:55 -03:00
diegosouzapw
1e652869d7 feat(api): add /api/quota/preview dry-run route (B/F8) 2026-05-28 07:49:47 -03:00
diegosouzapw
043f3c412b feat(api): add /api/quota/plans CRUD routes (B/F8) 2026-05-28 07:49:38 -03:00
diegosouzapw
9dfea1e7ad feat(api): add /api/quota/pools CRUD + usage routes (B/F8) 2026-05-28 07:49:30 -03:00
diegosouzapw
0a398002bf Merge F10 (audit + E2E spec) into refactor/pages-v3-19 2026-05-28 07:47:30 -03:00
diegosouzapw
cd768b8902 chore(translator): F10 audit — add Playwright E2E spec for translator friendly redesign 2026-05-28 07:47:14 -03:00