Menu
Convert and extract text from PDFs, DOCX, images (OCR), and other document formats using the gateway's built-in document processing stack.
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
Manage the Trinity AGI auth service and RBAC system -- database schema, role hierarchy, permission management, JWT flow, audit logging, and API endpoints.
Sync files between the host machine and the running Trinity Docker containers — config, agent files, skills, and extensions.
Develop the Trinity AGI Flutter Web Shell -- understand the architecture, A2UI renderer, WebSocket models, dual-client pattern, design tokens, and build workflow.
Operate the Trinity AGI Kubernetes infrastructure -- Helm charts, multi-tenant per-user OpenClaw pods, Vault Agent Injector secrets, minikube dev, and production deployment.
Operate the OpenClaw Gateway that powers Trinity AGI — configure providers, manage sessions, use the CLI, and understand the WebSocket protocol.
Manage the Trinity AGI Docker Compose stack — start, stop, logs, rebuild frontend, and full deploy.
| name | doc-convert |
| description | Convert and extract text from PDFs, DOCX, images (OCR), and other document formats using the gateway's built-in document processing stack. |
| homepage | https://github.com/trinityagi/trinity |
| metadata | {"openclaw":{"emoji":"📑","requires":{"bins":["pdftotext","pandoc","tesseract"]}}} |
Extract text from PDFs, DOCX files, images (via OCR), and other document formats. All tools are pre-installed in the OpenClaw gateway container.
| Tool | Binary | Purpose |
|---|---|---|
| pdftotext | pdftotext (poppler-utils) | Extract text from PDF files |
| pdfinfo | pdfinfo (poppler-utils) | Get PDF metadata (page count, size, etc.) |
| pdfimages | pdfimages (poppler-utils) | Extract images from PDFs |
| pandoc | pandoc | Convert between formats (DOCX->text, HTML->md, etc.) |
| tesseract | tesseract | OCR - extract text from images |
| python3 | python3 | pdfplumber, PyPDF2, python-docx, Pillow, pytesseract |
| libreoffice | libreoffice | Convert DOCX/XLSX/PPTX to PDF |
| imagemagick | convert / identify | Image manipulation and format conversion |
pdftotext input.pdf - # stdout
pdftotext -layout input.pdf output.txt # preserve layout
pdftotext -f 1 -l 5 input.pdf - # pages 1-5 only
pdfinfo input.pdf
pandoc -f docx -t plain input.docx # stdout
pandoc -f docx -t markdown input.docx # as markdown
libreoffice --headless --convert-to pdf input.docx
tesseract image.png stdout # extract text from image
tesseract image.png output -l eng pdf # OCR to searchable PDF
tesseract image.png stdout -l chi_sim # Chinese OCR
pandoc -f html -t markdown input.html
# PDF with tables
import pdfplumber
with pdfplumber.open("file.pdf") as pdf:
for page in pdf.pages:
print(page.extract_text())
for table in page.extract_tables():
print(table)
# DOCX
from docx import Document
doc = Document("file.docx")
for para in doc.paragraphs:
print(para.text)
# Image OCR via Python
from PIL import Image
import pytesseract
text = pytesseract.image_to_string(Image.open("image.png"))
print(text)
| Format | Read | Convert To |
|---|---|---|
| pdftotext, pdfplumber, PyPDF2 | text, markdown, images | |
| DOCX | pandoc, python-docx | text, markdown, PDF |
| HTML | pandoc | text, markdown, PDF |
| Images (PNG/JPG/TIFF/BMP) | tesseract (OCR) | text, searchable PDF |
| XLSX/PPTX | libreoffice | PDF, then text |
| RTF | pandoc | text, markdown |
| EPUB | pandoc | text, markdown |
| ODT | pandoc, libreoffice | text, PDF |
tesseract or pdfplumber with image extraction.pdftotext is fastest for text-based PDFs. Use pdfplumber when you need table extraction.pandoc handles format conversion between most document types.pdftotext -f 1 -l 10) to avoid memory issues.