jxxghp 221eb21694 refine internal middleware llm usage for streaming agents ...

Use a non-streaming model for middleware-only calls so internal outputs do not leak into user streams and model-based middleware stays consistent.

2026-04-27 06:55:41 +08:00

..

agent

refine internal middleware llm usage for streaming agents

2026-04-27 06:55:41 +08:00

api

refactor: remove legacy LLM_DISABLE_THINKING and LLM_REASONING_EFFORT config, unify thinking_level handling

2026-04-25 10:42:03 +08:00

chain

feat: add agent session usage status reporting

2026-04-26 08:19:05 +08:00

core

refactor: remove legacy LLM_DISABLE_THINKING and LLM_REASONING_EFFORT config, unify thinking_level handling

2026-04-25 10:42:03 +08:00

db

fix local db initialization model registration

2026-04-16 17:05:57 +08:00

helper

fix: improve DeepSeek reasoning_content payload handling and update langchain dependencies

2026-04-25 12:46:21 +08:00

modules

feat: add skills marketplace management

2026-04-22 14:55:00 +08:00

plugins

…

schemas

feat: add ai-compatible API endpoints

2026-04-22 17:21:43 +08:00

startup

…

utils

feat: normalize internal system user ID in notification dispatch

2026-04-21 14:32:14 +08:00

workflow

…

__init__.py

…

cli.py

feat: improve local CLI startup management

2026-04-21 11:26:56 +08:00

command.py

feat: add agent session usage status reporting

2026-04-26 08:19:05 +08:00

factory.py

…

log.py

feat: improve local CLI startup management

2026-04-21 11:26:56 +08:00

main.py

feat: improve local CLI startup management

2026-04-21 11:26:56 +08:00

monitor.py

…

scheduler.py

…