Menu
Upload, vision-parse, and persist a document with multi-level summaries.
Install with Codex or Claude Copy this prompt, paste it into Codex, Claude, or another assistant, and let it review the skill page and install it for you.
Based on SOC occupation classification
Generate a Word document from ingested source documents using the agnostic Document tree (CommonMark/Pandoc-aligned)
Generate a single Document (Word/PowerPoint) from all source documents in a collection, with attributions back to each source
Generate a PDF from ingested source documents using the agnostic Document tree
Generate a PowerPoint deck from ingested source documents using the agnostic Document tree (parametrised by slide_count, style, audience, tone)
Generate an Excel workbook (.xlsx) from ingested source documents โ tabular with charts, formulas, KPI cells
Find PII (emails, phones, SSN, credit cards, IBANs, names, addresses, DOB) in an ingested document; return redaction plan
| name | document-ingest |
| description | Upload, vision-parse, and persist a document with multi-level summaries. |
| inputs | [{"name":"file_path","type":"string","required":true,"description":"Local or remote path to the source file."},{"name":"filename","type":"string","required":true,"description":"Human-readable filename stored in the Document record."},{"name":"collection_id","type":"string","required":false,"description":"Optional folder/collection to assign the document to on creation."}] |
| outputs | [{"name":"document_id","type":"string","description":"UUID of the newly created Document row."},{"name":"page_count","type":"integer","description":"Number of pages parsed from the document."}] |
| tools | ["vision-parse","db-query"] |
Pick this skill when the user uploads a file or references a path that has not yet been imported into the system. Use it before any classify, summarize, or extract operation because those skills assume the document already exists in the store.
vision-parse (DOCX โ PDF โ per-page images).Document row, one Chapter row per chapter, one Section per section,
and one Page per page to the DB via db-query.Chunk rows; optionally embed and store vectors.Document.summary, Chapter.summary, and Section.summary.Emits one ExecutionHistory row with event_type=DOCUMENT_IMPORTED and actor_type=agent
after the Document row is committed.