-
Notifications
You must be signed in to change notification settings - Fork 0
Description
Is your feature request related to a problem? Please describe.
Architects aho are transitioning to docs-as-code workflow possess large, legacy documents from external tools (e.g., EA, Word, ReqIF, or enterprise architect tools). It is hard to manually and tediously convert these massive documents into reStructuredText (RST) format, extract requirements into sphinx-needs directives, and convert embedded diagrams (like those in Visio or Word) into formats like PlantUML or Mermaid. This manual process is time-consuming, highly error-prone, and may act as a significant bottleneck to adopting the docs-as-code approach.
Describe the solution you'd like
Specialized Transition Agent tool that should automate the conversion process by reading content from various legacy document formats (e.g., Word, ReqIF).
The agent should be:
- Schema-Aware: It must understand the defined sphinx-needs schema model to intelligently structure the output.
- Content Translation: It should create a draft RST document that includes:
- One-to-one translation of the source document's text into correctly formatted RST.
- Conversion of embedded or referenced diagrams into PlantUML or Mermaid syntax.
- A proposal for new needs directives, automatically extracted and created based on the source document's content (e.g., based on headings, numbering, or keywords).
The final output should be a draft document that serves as the starting point for users to finalize their transition to docs-as-code.
Describe alternatives you've considered
Do everything manually.
Additional context
...