mirror of
https://github.com/danny-avila/LibreChat.git
synced 2026-09-13 10:11:25 +00:00
34e2ea9319
The provisioning service and its ON_TOOL_EXECUTE callback were 600 lines of implementation under /api, which the workspace rules reserve for thin wrappers, so none of it was type checked despite its TypeScript callers. Both now live in packages/api. Storage strategies, vector upload, credential lookup, and the file model are api-workspace concerns, so they are injected rather than imported, and what remains on the JS side is 31 lines of wiring. The service is built on first use for the same reason its axios instance is: this module is reachable from the OpenAI-compatible controllers, whose suites partially mock the package, and a load-time call would throw before any provisioning is requested. Tests move with the logic and improve in the process: they now exercise the real service with injected fakes instead of mocking the package that contains it. |
||
|---|---|---|
| .. | ||
| askUserQuestion.e2e.spec.js | ||
| callbacks.spec.js | ||
| client.contextMetadata.spec.js | ||
| client.eventActorPause.spec.js | ||
| client.memory.spec.js | ||
| client.steerWiring.spec.js | ||
| client.subagentUsage.spec.js | ||
| client.transactions.spec.js | ||
| hitlCheckpoint.e2e.spec.js | ||
| jobReplacement.spec.js | ||
| modelEndHandler.spec.js | ||
| openai.spec.js | ||
| protocol.spec.js | ||
| request.partialDisconnect.spec.js | ||
| request.resumeMetadata.spec.js | ||
| responses.unit.spec.js | ||
| resume.spec.js | ||
| steer.spec.js | ||
| usageEvents.integration.spec.js | ||
| usageEvents.live.spec.js | ||
| v1.duplicate-actions.spec.js | ||
| v1.spec.js | ||