docproc --file input.pdf -o output.md
# With explicit config
docproc --file input.pdf -o output.md --config docproc.yaml
# Verbose
docproc --file slides.pptx -o slides.md -vPDF, DOCX, PPTX, XLSX. Output must be a .md file (-o output.md).
docproc init-config --env .env
# Writes ~/.config/docproc/docproc.yml from your .env (AI keys, etc.)docproc completions bash # or zsh
# Source the output in your shell to get completions for --file, -o, --config- CONFIGURATION.md — Config schema, AI providers, ingest options
- AZURE_SETUP.md — Azure OpenAI and Azure AI Vision setup
For the document workspace (upload, RAG chat, notes, assessments), see demo/README.md. The demo uses Docker Compose for infrastructure (PostgreSQL, LocalStack, RabbitMQ) and runs the Go API and React frontend on the host. Document processing is done by the docproc CLI invoked from the Go worker.