Labels: ready-to-code
Depends on: Issue 2
Create the Kagenti provider module as an independent createBackendModule with KagentiProvider, A2A protocol integration, and all provider caches using cacheService.
Tasks
From openspec/changes/pluggable-ai-platform-architecture/tasks.md sections 2 and 3:
- 2.3 Create
boost-backend-module-kagenti package
- 2.4 Implement
KagentiProvider and KagentiProviderFactory
- 2.5 Ensure no cross-provider imports
- 3.3 Implement agent card cache via
cache.withOptions({ defaultTtl: '5m' })
- 3.4 Implement Keycloak token cache
- 3.7 Implement session maps via cacheService
- 3b.1 Define Kagenti-specific types in module only
Specifications
openspec/changes/pluggable-ai-platform-architecture/specs/provider-packaging/spec.md — Kagenti module scenarios
openspec/changes/pluggable-ai-platform-architecture/design.md — Decision 2 (modules not plugins), Decision 3 (cacheService)
Labels:
ready-to-codeDepends on: Issue 2
Create the Kagenti provider module as an independent
createBackendModulewithKagentiProvider, A2A protocol integration, and all provider caches usingcacheService.Tasks
From
openspec/changes/pluggable-ai-platform-architecture/tasks.mdsections 2 and 3:boost-backend-module-kagentipackageKagentiProviderandKagentiProviderFactorycache.withOptions({ defaultTtl: '5m' })Specifications
openspec/changes/pluggable-ai-platform-architecture/specs/provider-packaging/spec.md— Kagenti module scenariosopenspec/changes/pluggable-ai-platform-architecture/design.md— Decision 2 (modules not plugins), Decision 3 (cacheService)