ragleap-graph v0.6.1 release notes

Added

GraphRetriever.retrieve()'s graph_context.related_documents entries now include an already_in_vector_results boolean flag. Closes a known limitation: a document could genuinely surface through both vector search (as a chunk) and graph traversal (as a related document), with no way for the caller to know about the overlap.

Deliberately flags rather than removes overlapping entries — removing them would silently discard real graph-relationship context (matched entities, graph score) that's still useful even for a document also present in the chunk list. The caller decides what to do with the flag.

Verified

Regression test added first using the existing FakeRag/FakeGraph test doubles (no live services needed — GraphRetriever has no I/O of its own), confirmed failing (KeyError) before the fix, passing after.

Full suite: 81 passed, 1 skipped (unrelated), zero regressions.

PyPI: https://pypi.org/project/ragleap-graph/0.6.1/
Full CHANGELOG: https://github.com/antonyrag/ragleap-core/blob/main/packages/ragleap-graph/CHANGELOG.md