From a4ae0c494c517b4ecb8c585b874047e6f15861e9 Mon Sep 17 00:00:00 2001 From: Adarsh Prashar Date: Sun, 14 Jun 2026 21:19:35 +0530 Subject: [PATCH] docs(changelog): remove duplicate streaming-proxy entry MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit The Unreleased section listed the streaming proxy twice — the original OpenAI-only entry and the later one covering both /v1/chat/completions and /v1/messages. Keep the both-endpoints entry; drop the superseded duplicate. --- CHANGELOG.md | 8 -------- 1 file changed, 8 deletions(-) diff --git a/CHANGELOG.md b/CHANGELOG.md index 1c43cde..aa5d66e 100644 --- a/CHANGELOG.md +++ b/CHANGELOG.md @@ -50,14 +50,6 @@ surface is governed by [`COMPATIBILITY.md`](COMPATIBILITY.md). `gate_tools=True` to route `CBEventType.FUNCTION_CALL` through the approval gate. `llama-index-core` is lazily imported, so the SDK stays dependency-free; pinned to the `llama-index-core` >= 0.10 callback protocol. -- **Streaming proxy.** `POST /v1/chat/completions` now supports `stream:true`: the - budget is enforced before the stream opens, the OpenAI provider's SSE is forwarded - to the client verbatim (authentic chunks, no translation) while token usage is - metered from the final usage chunk, and the run's context — time budget, kill - switch, or client disconnect — cuts a live stream. Dollar/token budgets are - checked pre-stream and recorded after (so the next call is refused if it went - over). A provider without streaming, and the Anthropic `/v1/messages` endpoint, - return a clear 501 rather than silently buffering. - **Prometheus `/metrics` endpoint.** Scrape the daemon's own state: governed runs by status, halted runs by halt reason, total spend in dollars and tokens, priced model calls, and the pending-approval queue depth. Plain Prometheus text