Full deployment history for this project.
🛡️ Sentinel: Protect evaluate API endpoints with withAuth (#387)
🛡️ Sentinel: Protect evaluate API endpoints with withAuth Co-authored-by: google-labs-jules[bot] <161369871+google-labs-jules[bot]@users.noreply.github.com>
⚡ Bolt: Add request-level caching to GraderService (#386)
⚡ Bolt: Add request-level caching to GraderService Add a short-lived request-level in-memory cache (1000ms TTL) for `GraderService` read operations (`list`, `getById`, `getLatestVersion`). Automatically clear the cache on state mutations (`create`, `update`, `delete`, `createVersion`). This deduplicates repeated SQL queries over serverless DB connections during batch evaluation execution. Co-authored-by: google-labs-jules[bot] <161369871+google-labs-jules[bot]@users.noreply.github.com>
🎨 Palette: Enhance dashboard status badge accessibility (#385)
🎨 Palette: Enhance dashboard status badge accessibility Add role="img", aria-label, and title attributes to the Badge component in components/dashboard/badge.tsx for clear screen reader announcements and hover tooltips across dashboard cards and metrics. Co-authored-by: google-labs-jules[bot] <161369871+google-labs-jules[bot]@users.noreply.github.com>
test(evals): add comprehensive unit tests for traceContextFromRequest (#363)
🧹 remove unused RegistrySource and RegistryStatus imports in registry page (#364)
fix(web4): remove unused EvalsFingerprintType import in fingerprints.ts (#366)
⚡ bulk insert evaluation case results in run orchestrator (#380)
🧹 Refactor long bulkUpsertCapabilities function (#345)
⚡ Bolt: Align evidenceIds to checks index in provider run adapters (#347)
🧹 Refactor EvaluationLauncher into modular sub-components (#342)
🧪 Add unit tests for redactEvidenceText security redaction utility (#343)
🧪 Add unit tests for buildJsonLd and discovery endpoints (#344)
🎨 Palette: improve EvaluationLauncher accessibility and error feedback (#340)
perf(registry): parallelize service and capability soft deletes in sync Registry (#341)
fix(tests): handle optional fingerprintContext in correlation tests (#339)
⚡ Bolt: Bulk insert eval_run_cases in chunked orchestrator execution (#308)
⚡ Bolt: Map lookup optimization in RegressionService.detectRegressions (#326)
🧪 test(coverage): add NetworkCoverageGraph UI and calculation tests (#328)
test: add missing branch coverage for correlation context & headers (#332)
test: add comprehensive error handling and bulk ops tests for syncRegistry (#333)
a11y(runs): add ARIA progressbar semantics to score bars (#306)
fix(security): prevent SSRF bypass via embedded non-standard IPv4 in IPv6 notation (#307)
⚡ Bolt: Optimize suite test cases query in API routes (#305)
🎨 Palette: Enhance CopyButton layout resilience and transition state (#286)
fix(security): prevent SSRF bypass via POSIX shortened IPv4 notation (#302)
⚡ Bolt: Parallelize suite validation and case retrieval in RunOrchestrator.execute (#304)
⚡ Bolt: Optimize bulkUpsertCapabilities query waterfalls and update batching (#297)
Merge branch 'main' into palette/a11y-table-headers-2971129657162063939
Merge branch 'main' into jules-15472953114793677214-c958f945
Merge branch 'main' into bolt-release-gate-service-caching-14185807193815037532
Merge branch 'main' into bolt-consolidate-suite-lookups-17064952329711643281
Merge branch 'main' into bolt-parallel-eval-case-creation-9925164005846661263
Merge branch 'main' into jules-palette-table-headers-a11y-2260142099760902119
chore: stack PR #194
chore: validate consolidated upgrades and platform auth
Verify UI button route integrity (no 404 regressions) and document findings Agent-Logs-Url: https://github.com/mbarbine/platphorm-evals/sessions/351b9647-cadf-4d6c-87d3-1158d386ee48 Co-authored-by: mbarbine <3211492+mbarbine@users.noreply.github.com>
fix: use real persisted data for homepage network coverage Agent-Logs-Url: https://github.com/mbarbine/platphorm-evals/sessions/53d9f985-a1f9-43ef-a9f1-3ea2fcb8f994 Co-authored-by: mbarbine <3211492+mbarbine@users.noreply.github.com>
Extract extractCount helper in homepage to reduce cast repetition Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com> Co-authored-by: mbarbine <3211492+mbarbine@users.noreply.github.com>
feat: service probing, registry routes, expanded OpenAPI, health DB status, public MCP GET - lib/registry/service-probing.ts: probes canonical PlatPhorm Network services (health, docs, mcp, llms.txt, sitemap, rss) with 10s timeout, SSRF guard, capability extraction, and parallel probeAllServices() - app/api/v1/registry/services/route.ts: GET /api/v1/registry/services (no auth, paginated) - app/api/v1/registry/probe/route.ts: POST /api/v1/registry/probe (auth required, read-only) - lib/discovery/openapi.ts: document all API endpoints with tags, params, and response schemas - app/api/health/route.ts: DB availability check, serviceCount, boolean integration flags - app/api/mcp/route.ts: make GET public (no auth), keep POST with auth, add ping method Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com> Co-authored-by: mbarbine <3211492+mbarbine@users.noreply.github.com>