llm_extraction integration by adding the bridge into candidate graph drafts

README.md

@@ -117,8 +117,10 @@ python -m pip install -e ../llm-connect
 
 The integration accepts any `llm-connect` style adapter with
 `execute_prompt(prompt, config)` and parses strict JSON candidate drafts from
-model responses. Tests use a fake adapter, so the default test suite does not
-call external providers.
+model responses. Parsed drafts can be mapped into reviewable candidate graph
+entries while preserving source paths where they match observed facts or
+content chunks. Tests use fake adapters, so the default test suite does not call
+external providers.
 
 ## Agent-Facing Endpoints