R
Use R Markdown
Write reproducible R reports with Markdown prose, R chunks, plots, and rendered output.
What R Markdown is for
R Markdown combines prose, R code, output, and document rendering in one file. Use it for reports, statistical notebooks, teaching materials, and reproducible analysis that should render to HTML, PDF, or other formats.
Create and render
- Create a
.Rmdfile. - Put document settings in the YAML header.
- Write Markdown prose.
- Add R chunks for computations, plots, tables, and checks.
- Render the document and inspect the output.
Run chunks incrementally while developing. If a full render fails, rerun the failing chunk in a fresh session and check package availability in the project environment.
Ask Agent to repair a render failure
A failed R Markdown render can offer Fix with Agent... below the error output when AI help is available. Use that control to carry the render failure into an agent conversation.
- Render the document and inspect the error output in the Build Log.
- Choose Fix with Agent.... If you want debugging guidance first and the option is available, choose Ask Agent for a Hint....
- Review the error context in the confirmation and select a recent agent session when shown.
- Confirm with Send to Agent or Ask Agent, then follow the request in the agent chat.
- Inspect any changed source and render the document again to check the result.
The request includes the render error and build command. When the log identifies a source-line range, that range is also supplied. The agent is directed to read the live document before editing. Ordinary informational output from a successful render does not trigger this failed-build shortcut.
Reproducibility
Keep package setup, data paths, and rendering commands explicit. For classes or shared research projects, prefer a runtime image or setup script so the R environment is reproducible across projects.