Fran Jurmanović df4ca0bf76 Track cached prompt tokens through to Langfuse ...

OpenAI's chat completions response includes prompt_tokens_details.
cached_tokens when prompt caching kicks in - previously dropped
entirely. Now extracted in LlmClient and forwarded to Langfuse as the
flat cache_read_input_tokens usage key.

Used the flat key rather than OpenAI's nested prompt_tokens_details
shape: the installed @langfuse/tracing SDK's usageDetails type is
Record<string, number> | OpenAiUsage (a 3-field camelCase shape, no
_details nesting) even though the wire schema in @langfuse/core
supports nesting - a gap in the SDK's public types, not something
worth fighting with a cast.

Verified: full suite (82/82) passing, tsc/eslint clean.

2026-08-12 23:15:42 +02:00

..

LlmClient.test.ts

Track cached prompt tokens through to Langfuse

2026-08-12 23:15:42 +02:00

LlmPipeline.test.ts

Group pipeline runs into Langfuse sessions, tagged by trigger source

2026-08-12 22:51:17 +02:00

LlmTool.test.ts

Fix final review findings: key leak, model fallback, error resilience, docs

2026-08-12 21:40:09 +02:00

SchemaLlmTool.test.ts

Add SchemaLlmTool and TextLlmTool

2026-08-12 20:54:10 +02:00

TextLlmTool.test.ts

Add SchemaLlmTool and TextLlmTool

2026-08-12 20:54:10 +02:00