Zhou Kai e4a7a04719

feat(subagents): add isolated date-only context (#4797) ...

* feat(subagents): inject date-only runtime context

* refactor(middleware): deduplicate date reminder formatting

2026-08-13 23:37:46 +08:00

..

__init__.py

feat: add create_deerflow_agent SDK entry point (Phase 1) (#1203)

2026-03-29 15:31:18 +08:00

_bounded_dict.py

fix(subagents): unify guardrail caps on additive stop_reason + add token_budget (#3875 Phase 2) (#3980)

2026-07-08 22:26:06 +08:00

AGENTS.md

feat(subagents): add isolated date-only context (#4797)

2026-08-13 23:37:46 +08:00

clarification_middleware.py

fix(clarification): normalize dict options (#4527)

2026-07-28 19:43:56 +08:00

configured_extensions.py

feat(agent): support config-declared lead middlewares (#3964)

2026-07-21 09:39:44 +08:00

dangling_tool_call_middleware.py

fix(middleware): skip raw tool-call fallback when invalid view carries the same call (#4693)

2026-08-08 21:07:55 +08:00

deferred_tool_filter_middleware.py

fix(tool-search): reliably hide deferred MCP schemas by removing the ContextVar (closures + graph state) (#3342)

2026-06-02 22:43:22 +08:00

delegation_ledger.py

fix(subagents): unify guardrail caps on additive stop_reason + add token_budget (#3875 Phase 2) (#3980)

2026-07-08 22:26:06 +08:00

durable_context_middleware.py

feat(agent): Add subagent total delegation cap (#4115)

2026-07-13 19:26:07 +08:00

dynamic_context_middleware.py

feat(subagents): add isolated date-only context (#4797)

2026-08-13 23:37:46 +08:00

input_sanitization_middleware.py

fix(middleware): sanitize bare string blocks in list-form user content (#4668)

2026-08-05 08:06:01 +08:00

llm_error_handling_middleware.py

fix(streaming): expose custom events to astream_events (#4403)

2026-07-23 22:56:12 +08:00

loop_detection_middleware.py

fix(loop-detection): clear the per-tool frequency counter on evict/reset (#4295)

2026-07-19 17:13:44 +08:00

mcp_routing_middleware.py

feat(mcp): auto-promote deferred MCP tools from routing hints (#4019)

2026-07-10 07:54:36 +08:00

memory_middleware.py

feat(memory): add mem0 HTTP memory backend (#4528)

2026-07-29 07:11:20 +08:00

model_length_finish_reason_middleware.py

fix: surface length-capped model responses (#4309)

2026-07-26 14:43:08 +08:00

model_length_termination_detectors.py

fix: surface length-capped model responses (#4309)

2026-07-26 14:43:08 +08:00

read_before_write_middleware.py

feat(middleware): add structured tool result meta and tool-progress state machine (#3601)

2026-07-06 20:57:41 +08:00

safety_finish_reason_middleware.py

fix(streaming): expose custom events to astream_events (#4403)

2026-07-23 22:56:12 +08:00

safety_termination_detectors.py

fix(runtime): suppress tool execution when provider safety-terminates with tool_calls (#3035)

2026-05-22 21:20:28 +08:00

sandbox_audit_middleware.py

fix(sandbox): judge command substitution by position in audit middleware (#4623)

2026-08-01 21:44:34 +08:00

skill_activation_middleware.py

fix(run): add run event stream contract (#4342)

2026-07-23 21:33:57 +08:00

skill_context.py

feat: emit structured runtime metadata (follow-up#3887) (#3906)

2026-07-04 11:27:19 +08:00

skill_tool_policy_middleware.py

fix(subagents): isolate callbacks and activate skills lazily (#4497)

2026-07-28 23:29:14 +08:00

subagent_limit_middleware.py

fix(subagents): clamp subagent limit consistently with MIN_SUBAGENT_LIMIT (#4081)

2026-07-24 21:56:11 +08:00

summarization_middleware.py

feat(extensions): observe task lifecycle and system model calls (#4684)

2026-08-11 16:33:22 +08:00

system_message_coalescing_middleware.py

fix(agents): coalesce SystemMessages before LLM request (#3711)

2026-06-24 07:28:17 +08:00

terminal_response_middleware.py

fix: ensure visible response after tool runs (#4033)

2026-07-11 15:53:14 +08:00

thread_data_middleware.py

fix(runtime): honor LangGraph Server identity for user-scoped data (#4538)

2026-07-28 22:59:14 +08:00

title_middleware.py

feat(extensions): observe task lifecycle and system model calls (#4684)

2026-08-11 16:33:22 +08:00

todo_middleware.py

fix(todo-middleware): call super in wrap_model_call to restore write_todos prompt injection (#4714) (#4735)

2026-08-11 11:41:31 +08:00

token_budget_middleware.py

fix: read run stop_reason from runtime context (#4188)

2026-07-16 08:19:52 +08:00

token_usage_middleware.py

fix(subagents): isolate background tasks from reused tool call IDs (#4758)

2026-08-12 09:25:05 +08:00

tool_call_metadata.py

fix(middleware): sync raw tool call metadata (#2757)

2026-05-08 10:08:53 +08:00

tool_error_handling_middleware.py

feat(subagents): add isolated date-only context (#4797)

2026-08-13 23:37:46 +08:00

tool_output_budget_middleware.py

feat(tool):Add structured synopsis for oversized tool output previews (#3377)

2026-07-16 16:41:04 +08:00

tool_output_synopsis.py

feat(tool):Add structured synopsis for oversized tool output previews (#3377)

2026-07-16 16:41:04 +08:00

tool_progress_middleware.py

feat(middleware): add structured tool result meta and tool-progress state machine (#3601)

2026-07-06 20:57:41 +08:00

tool_result_meta.py

fix(agents): classify web_fetch error pages as errors, not successful evidence (#4314)

2026-07-20 23:05:47 +08:00

tool_result_sanitization_middleware.py

fix(security): neutralize prompt-injection tags in web_capture results (#4099)

2026-07-12 11:24:24 +08:00

uploads_middleware.py

fix(runtime): honor LangGraph Server identity for user-scoped data (#4538)

2026-07-28 22:59:14 +08:00

view_image_middleware.py

fix(backend): remove transient image context after model calls (#4267)

2026-07-22 08:41:50 +08:00