TeXRA

Appearance

Working with Figures ​

TeXRA allows AI agents to analyze, reference, and generate figures within your documents.

Quick Task: Add a Figure Caption ​

  1. Select the polish agent from the dropdown
  2. Choose a vision-capable model (e.g., gemini25p, gpt4o)
  3. Select your figure in the Media section
  4. Enter instruction: "Write a detailed caption for this figure"
  5. Click Execute

The Media Section ​

The main TeXRA panel includes a "Media" section for managing figure files:

  • Dropdown: Select a primary media file
  • Multiple Toggle: Expand to select multiple files
  • Auto Extract Dropdown: Configure automatic figure extraction

Supported File Types ​

Configurable via texra.files.included.mediaExtensions:

  • Images: .png, .jpeg, .jpg, .gif, .heic, .heif, .webp
  • Documents: .pdf (native or converted to images)
  • Audio (experimental): .wav, .m4a, .mp3, .aiff, .aac, .ogg, .flac

PDFs are processed natively when supported (Anthropic/Gemini/OpenAI). Otherwise, TeXRA uses GraphicsMagick/ImageMagick + Ghostscript for conversion.

Clipboard Images ​

Paste images directly into the instruction area:

  1. Copy any image to clipboard
  2. Paste with Ctrl/Cmd+V
  3. Image is saved and referenced as [pasted_timestamp_hash.ext]
  4. Media Files list updates automatically

Pasted images are stored temporarily and cleaned up after 3 days.

Automatic Figure Extraction ​

Enable via the Auto-extract dropdown near the Media label:

  • Figures: Extracts images from \includegraphics commands
  • TikZ Figures: Extracts tikzpicture environments as .tikz files

Figure Extraction Tools ​

Tool-use agents can extract figures programmatically:

extract_figures ​

Scans for \includegraphics and returns referenced files (max 20 attachments):

json
{
  "name": "extract_figures",
  "arguments": { "texPath": "paper/main.tex" }
}

extract_tikz_figures ​

Extracts and optionally compiles TikZ environments (max 12 PDFs):

json
{
  "name": "extract_tikz_figures",
  "arguments": { "texPath": "paper/main.tex", "compile": true }
}

extract_bib_entries ​

Retrieves BibTeX records for citations.

Using Media Files ​

When you provide media files:

  • Vision models (GPT-4o, Gemini): Images are encoded and included with the prompt
  • Audio models: Audio files are uploaded for transcription
  • Non-multimodal models: Filenames provide context

Common use cases:

  • Write captions for images (polish agent)
  • Verify text matches figures (correct agent)
  • Generate text from images/PDFs (ocr agent)
  • Transcribe audio (transcribe_audio agent)

For TikZ-specific workflows, see TikZ Figures.