AI document & PDF skills for agents
Document work is format plus judgment: consistent headings, tables that survive export, plain-language policy text, exec summaries that keep the numbers. These skills teach both layers.
Includes the document-engineering skills for docx and pdf — with scan verdicts and licenses.
9 skills · 1 benchmarked with Lift · sorted by measured performance, then Trust.
Records decisions and documentation, including architecture decision records and code documentation, to provide context for future engineers
Audits README, SPEC, and PRODUCT docs against recent git history to detect drift and opens minimal PR-ready accuracy fixes.
Maintains a narrowly scoped CONTEXT.md capturing stable domain language and routes hard-to-reverse decisions to the existing architecture-de
Routes broad documentation-governance requests to the right specialized skill and runs a read-only audit of documentation roles, links, ADR
Creates and keeps consistent investor-facing fundraising materials like pitch decks, one-pagers, memos, financial models, and accelerator ap
Creates, edits, reads, and formats Word .docx/.dotx files, including tracked changes, comments, tables, and templates.
Handles all PDF operations including reading, extracting text and tables, merging, splitting, rotating, watermarking, creating, filling form
Generates structured medical case reports in SOAP or narrative format with consistent vitals, labs, and evidence-based management plans for
Generates a two-page digital e-guide preview with a styled cover and inside spread in a lifestyle creator tone.
FAQ
What is an AI document generator skill?
A layout-and-voice pack — structure, tone, formatting rules for Word/PDF output — plus the file-generation workflow itself.
Can agents chat with PDFs using these?
Document-Q&A skills teach extraction discipline (what's actually on the page vs inferred) that makes PDF answers citable.