# `Feline.Processors.SentenceAggregator`
[🔗](https://github.com/dimamik/feline/blob/main/lib/feline/processors/sentence_aggregator.ex#L1)

Buffers LLM token stream and emits complete sentences.

Collects `LLMTextFrame` chunks until a sentence boundary (`.`, `!`, `?`)
is found, then pushes each complete sentence as a `TextFrame`. Remaining
text is flushed on `LLMFullResponseEndFrame` or `EndFrame`.

# `child_spec`

---

*Consult [api-reference.md](api-reference.md) for complete listing*