A library funding agent needs grant PDF narratives as markdown so it can reason over sections cleanly. What should document analyzers produce?

A library funding agent that must reason over grant PDF narratives needs document analyzers to produce markdown conversions of those narratives. Markdown preserves headings, sections, and lists so the agent can address need statements, budgets, and outcomes as structured text rather than opaque PDF layout. Markdown summaries of the grant PDF narratives condense and rewrite the content rather than converting full sections, so the agent loses complete need statements, budget details, and outcomes needed for clean section-aware reasoning. Markdown embeddings of the grant PDF narratives turn the sections into vector representations for search rather than readable markdown text, so the agent cannot directly reason over need statements, budgets, and outcomes as structured sections. Markdown translations of the grant PDF narratives alter wording into another language rather than faithfully converting the original sections, so the agent cannot reason over the authoritative need statements, budgets, and outcomes. Exam caveat: grant-narrative reasoning for municipal funding agents depends on markdown conversion from the PDFs—not markdown summaries, markdown embeddings, or markdown translations. Operational check: convert sample grant narratives to markdown with section headings intact; have the funding agent answer section-specific questions with citations back to those headings; compare answers to the source PDF; reject markdown-summary-only, markdown-embedding-only, or markdown-translation-only analyzer outputs for this agent path before award recommendations ship. Keep original PDF page anchors alongside markdown headings so reviewers can jump from an agent citation to the exact grant page during funding committee review.