KunN-21andCursor 7afaecd617 fix(kimchi): strip reasoning_content echo to bound multi-turn input tokens ...

Clients echo full message history each turn including reasoning_content,
which the Kimchi OpenAI gateway counts as input tokens. Multi-turn convos
balloon to 100k+ tokens and the model returns empty content.

KimchiExecutor.transformRequest now strips reasoning_content from assistant
messages when it exceeds an 8-char threshold, preserving the 1-char
placeholder injectReasoningContent sets and keeping content intact.

Co-authored-by: Cursor <cursoragent@cursor.com>

2026-07-03 10:58:31 +07:00

..

config

fix(gemini): support native TTS generateContent endpoint

2026-06-26 10:50:13 +07:00

executors

fix(kimchi): strip reasoning_content echo to bound multi-turn input tokens

2026-07-03 10:58:31 +07:00

handlers

feat(kimchi): add Kimchi OAuth provider support

2026-06-29 15:29:17 +07:00

providers

fix(kiro): add Claude Sonnet 5 model support (#2264)

2026-07-03 10:54:15 +07:00

rtk

fix(headroom): skip unsafe responses tool history (#2132)

2026-06-29 15:59:21 +07:00

services

feat(clinepass): add ClinePass provider support

2026-07-03 10:53:41 +07:00

shared

refactor(open-sse): registry consolidation + DRY media/oauth/adhoc cleanup

2026-06-14 13:15:48 +07:00

transformer

fix(codex): harden streaming timeouts + Responses terminal events

2026-06-06 16:07:43 +07:00

translator

fix(antigravity): preserve Claude tool delta index (#2223)

2026-07-01 09:36:43 +07:00

utils

fix(codebuddy-cn): strip empty tool_calls arrays to preserve reasoning

2026-07-01 09:39:31 +07:00

.npmignore

Update jsconfig.json and package.json to correct open-sse path references from relative to local directory.

2026-01-05 10:37:09 +07:00

AGENTS.md

feat(headroom): add proxy lifecycle management + dashboard UI

2026-06-20 10:09:50 +07:00

index.js

refactor(open-sse): translator DRY + schema enums, bug fixes, dead code cleanup

2026-06-14 18:49:38 +07:00