How do I summarize a PDF into a Word document via API?

​

Steps

1. Store your API key in DOCONDA_API_KEY and install the SDK (pip install doconda or npm install @doconda/sdk).
2. Upload the PDF (and any images you want inside) with files.upload.
3. Create the document with sources and download the Word file.

from pathlib import Path
from doconda import Doconda

doconda = Doconda()  # reads the key from DOCONDA_API_KEY
law = doconda.files.upload("housing-law.pdf")
photo = doconda.files.upload("facade.jpg")
doc = doconda.documents.create(
    format="docx",
    prompt="Summarize the main measures for a client",
    sources=[law["id"], photo["id"]],
)
Path("summary.docx").write_bytes(doconda.documents.download(doc["id"], "docx"))

import { readFile, writeFile } from "node:fs/promises"
import { Doconda } from "@doconda/sdk"

const doconda = new Doconda()
const law = await doconda.files.upload(await readFile("housing-law.pdf"), "housing-law.pdf")
const photo = await doconda.files.upload(await readFile("facade.jpg"), "facade.jpg")
const doc = await doconda.documents.create({
  format: "docx",
  prompt: "Summarize the main measures for a client",
  sources: [law.id, photo.id],
})
await writeFile("summary.docx", await doconda.documents.download(doc.id, "docx"))

​

What you get back

• An editable Word file with the summary, and its PDF. Images go inside, where they fit.
• Warnings in the report (documents.report(id)):
  • content.unsupported_fact: a figure, date or fact that isn’t in your request or your sources.
  • content.unsupported_citation · content.citation_to_review: a legal citation that doesn’t come from your sources.
  • source.selected: your sources were very long and the part about what you asked for was used.
  • source.instructions: a source tried to give orders; it was ignored.
• status: ready, ready_with_warnings, needs_review or failed (not charged).

​

Limits

• In sources: up to 5 documents (Word, PowerPoint, Excel, PDF, also .doc, .ppt and .xls, TXT or Markdown) and up to 10 images, together.
• Each file up to 20 MB.
• For only the summary’s PDF, use format: "pdf"; for a presentation, format: "pptx".

​

FAQ

​

Keep reading

• Create and SDK: create documents.
• Quickstart.