nx_library_extract.nx

nx_library_extract.nx -- SOVEREIGN render+extract half of the digital-library best-practices research rung (L-R1a). Reads the six raw HTTP responses fetched by nx_library_fetch, strips HTTP headers, renders HTML->text with the team's nx_html_to_text, saves the rendered text, and runs the team's deterministic ce_extract (number+unit fact claims) over each. Lives in _hdl_build/ so the import path reaches both nx_claim_extract (here) and nx_html_to_text (../). Division: FETCH + RENDER + EXTRACT all sovereign Nishi; the qualitative "which sentence states which best practice" reading is the model seat (cited verbatim in the sourced research doc). expect_exit: 0 license_tier: ORIGINAL