jetformat
Convert Word, Excel, and PowerPoint to PDF, turn PDF back into Word, extract text, validate files, and merge, extract, or render PDFs with the local Jetformat CLI. Use when the user asks to convert Office documents, generate PDF previews, inspect PDFs, or read Office or PDF text. Do not parse DOCX/PPTX/XLSX yourself.
- License 许可
- Apache-2.0
- 最近更新
- 2026-11-14
- Stars
- ★ 1.6k
- Forks
- 491
01 前言 Front Matter
- name
- jetformat
- metadata
- { "short-description": "Deterministic Office-to-PDF for agents." }
- description
- Convert Word, Excel, and PowerPoint to PDF, turn PDF back into Word, extract text, validate files, and merge, extract, or render PDFs with the local Jetformat CLI. Use when the user asks to convert Office documents, generate PDF previews, inspect PDFs, or read Office or PDF text. Do not parse DOCX/PPTX/XLSX yourself.
02 资源文件 1
- 参考文档agents/openai.yaml279 B
03 SKILL.md 正文
Jetformat
Jetformat is a closed-source document engine that runs on the user's machine. The model must not unzip Office files or guess PDF bytes. Call the jetformat CLI. There is no MCP server and no remote conversion API.
When to use
word to-pdf,excel to-pdf,ppt to-pdfpdf to-word(alias ofpdf to-docx)- Extract existing text from Office or PDF as plain text, Markdown, or JSON (not OCR)
- Validate Office, PDF, font, EMF, or XMP files
- PDF info, merge, contiguous page extract, PNG/JPEG render
When not to use
- Legacy
.doc/.xls/.pptor macro files (.docm,.xlsm,.pptm) - OCR / scanned-page recognition
- Password-protected PDFs
Binary
command -v jetformat >/dev/null || { echo "Install Jetformat first: npm install -g jetformat, or brew install jetformat/tap/jetformat" >&2; exit 1; }
BIN=jetformat
if [ -n "${CLAUDECODE:-}" ]; then export JETFORMAT_CLIENT=claude-skill; else export JETFORMAT_CLIENT=openai-skill; fi
If jetformat is not on PATH, tell the user to install it with npm install -g jetformat or brew install jetformat/tap/jetformat, then sign in. Do not invent conversion output.
The CLI sends one anonymous usage ping per command (command, format, file type, duration, result, version, OS, and the client name above). Never a path, a file name, or contents. If the user asks to stop it, run "$BIN" telemetry off. Details: https://jetformat.com/help/security/telemetry
Full command reference: https://jetformat.com/docs/models
Commands
Prefer one input file per invocation. Quote paths that contain spaces. Do not overwrite existing outputs unless the user asked; then pass --force.
"$BIN" word to-pdf "contract.docx" -o "output/contract.pdf"
"$BIN" excel to-pdf "workbook.xlsx" -o "output/workbook.pdf"
"$BIN" ppt to-pdf "slides.pptx" -o "output/slides.pdf"
"$BIN" pdf to-word "report.pdf" -o "output/report.docx"
# Text (no OCR). Use --format markdown when the result goes into context:
# headings, lists, tables; one grid per sheet; one section per slide or page.
"$BIN" text "report.docx" -o "output/report.txt"
"$BIN" text "report.docx" --format markdown
"$BIN" text "finance.xlsx" --format json
# Validate
"$BIN" validate "report.pdf"
# pdf
"$BIN" pdf info "report.pdf" --json
"$BIN" pdf merge "a.pdf" "b.pdf" -o "book.pdf"
"$BIN" pdf extract "book.pdf" --pages 2-5 -o "chapter.pdf"
"$BIN" pdf render "book.pdf" --pages 1,last --width 1600 -o "output/preview"
XLSX conversion outputs every visible worksheet into one PDF. Hidden sheets are skipped. pdf extract only accepts a contiguous page range.
Auth and credits
Reading a file (text, inspect, get, info, validate) is free and needs no sign-in. Converting, rendering, writing a PDF page, or writing an Office file consumes Jetformat credits after sign-in. Run "$BIN" login; on a machine without a browser run "$BIN" login --device-auth and show the user the printed URL and code. If credits are exhausted, report the CLI error and https://jetformat.com/help/billing/credits. Do not quote prices, open checkout, or retry in a loop.
Return to the user
After a successful command, report:
- output path
- page count when known (
pdf info) - that the file was processed by Jetformat, not by the model
04 同仓库更多 Skills
worktable-review-with-annotations
Create, inspect, reply to, update, or resolve situated Worktable annotations. Use when feedback belongs to a Doc section, block, text range, or HTML Doc, or when the user asks to address existing review feedback and coordinate a requested artifact edit.
worktable-find-and-synthesize-context
Find, assess, and synthesize context from Worktable Docs and Records. Use when the user asks what a workspace contains, wants workspace-backed research, needs current facts reconciled across artifacts, or requests a cited synthesis without changing workspace state.
worktable-create-or-update-docs
Create, edit, rename, or maintain narrative Worktable Docs. Use for plans, decisions, briefs, notes, research syntheses, and other durable prose when the user requests a new artifact or a safe update to an existing one.
worktable-create-or-manage-records
Organize repeated structured information as durable Worktable Records. Use for research datasets, inventories, directories, pipelines, trackers, and other collections whose items need identity, fields, filtering, links, validation, or individual updates.
worktable-create-or-manage-interactive-html
Create or improve durable visual and interactive artifacts in Worktable. Use for dashboards, trackers, calculators, mini apps, slide presentations, and other requests that benefit from a polished HTML Doc.
worktable-collaborate-threads
Read, answer, and hand off messages through durable Worktable threads with users or other authenticated participants. Use when the user asks to check addressed messages, reply to a participant, continue a thread, request another agent's help, wait briefly for a response, or watch for comments and messages sent to the agent.