What problem does it solve? Getting a batch of third-party research-report PDFs into a shared, searchable team library normally means manual uploads, duplicate files, and no provenance tracking. This skill bulk-imports PDFs into a Chronicle organization as deduplicated, provenance-recorded assets that the server extracts and indexes for role-filtered search. ## Core Features & Use Cases - Bulk PDF import: Registers each PDF as an org-scoped imported_report asset with sha256-based deduplication, so re-running a batch reports already-present files instead of duplicating them. - Provenance and access control: Records sha256, size, source filename, and batch label in asset_config, and scopes search visibility to the organization (or a team), refusing personal-scope and public uploads. - Server-side extraction and indexing: Finalizing the batch enqueues extraction (math-capable OCR for scanned PDFs, layout parsing with derived HTML/Markdown views when configured) and indexes content into role-filtered search. - Use Case: A research lead says "import this folder of 40 papers into our org's library" — the skill resolves the organization, uploads each PDF with dedup, and reports imported/duplicate/failed counts plus the extraction job id. ## Quick Start Import all PDFs in the folder ./reading-list into our Chronicle organization and tell me which ones were already imported.
