Charles YaoandClaude Opus 4.8 27be3de30a feat(dify-builder): wire real Edit cognition into LlmBuilderAgent ...

analyze_impact/propose_edit_plan/build_edit_intents now call
services.dify_builder.agent.edit.* via _model_or_none() instead of
delegating to the canned PlaceholderAgent. 9 of the Protocol's 14
Build/Edit/Fix methods now use real LLM cognition; only
generate_mock_inputs and propose_build_repair (returns []) remain
deferred pending the live-test step.

Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>

2026-08-29 14:00:58 -07:00

..

__init__.py

…

build.py

fix(dify-builder): stop fabricating llm completion_params, dedupe dataset_ids grounding

2026-08-29 12:15:08 -07:00

edit.py

fix(dify-builder): guard analyze_impact target_node_ids; drop ARG noqa in edit tests

2026-08-29 13:54:08 -07:00

fix.py

…

graph_translate.py

fix(dify-builder): prevent duplicate start node ids on multiple starts

2026-08-29 11:36:02 -07:00

llm_agent.py

feat(dify-builder): wire real Edit cognition into LlmBuilderAgent

2026-08-29 14:00:58 -07:00

llm.py

…

model_resolver.py

…

ping.py

…

resources.py

…