ProgressBoard

The ProgressBoard is TeXRA's central hub for monitoring agent execution, viewing detailed logs, and managing past runs. Think of it as the mission control center for your AI assistant – essential for understanding what's happening under the hood and troubleshooting when things go sideways.

Accessing the ProgressBoard

The ProgressBoard typically appears in the Panel area at the bottom of your VS Code window.

• Automatic: It often opens automatically when you execute an agent.
• Manual: If it's closed, you can open it via the Command Palette (Ctrl+Shift+P or Cmd+Shift+P) by searching for View: Show TeXRA ProgressBoard.

Layout Overview

The ProgressBoard interface is split into two main sections (usually side-by-side, but configurable):

1. Stream Tabs: A list on the side (often right) showing different agent runs (streams).
2. Content Area: The main area (often left) displaying the header and log details for the currently selected stream.

Stream Tabs Section

This section lists all the agent execution streams from your current VS Code session.

• Switching Streams: Click on a stream name (e.g., polish@sonnet37: paper.tex) to view its specific logs and status in the Content Area.
• Delete All: The Delete All button at the bottom permanently removes all streams and their logs from the ProgressBoard view for the current session.

Content Area

This area shows the details for the stream selected in the Stream Tabs section.

Header

The header provides a summary and actions for the selected stream:

• Stream Name: Displays the identifier of the current run (e.g., agent@model: inputFile).
• Status Indicator: A colored circle shows the current status:
  • Green (Running): The agent is actively processing.
  • Grey (Stopped): The agent finished successfully or was stopped manually before completion.
  • Red (Error): The agent encountered an error during execution.
  • Yellow (Ready/Initial): The view is ready, but no stream is active yet.
• Action Buttons:
  • Stop: (Enabled only when running) Requests the current agent task to interrupt. Note: The agent might finish its current step (e.g., an API call) before fully stopping.
  • Run Again: (Enabled when stopped/error) Re-executes the agent using the exact same configuration (files, model, instruction) as this run.
  • Restore State: (Enabled when stopped/error) Loads the configuration (files, model, instruction, tool settings) from this run back into the main TeXRA view in the sidebar, allowing you to modify it before running again.
  • Diff: (Enabled when stopped/error, requires specific output files) Runs latexdiff on the output files generated by this stream, comparing them against their corresponding base files. (Requires appropriate .tex output).
  • Pack: (Enabled when stopped/error) Moves the output files and logs associated with this stream into a versioned folder within the History directory. See File Management.
  • Clean: (Enabled when stopped/error) Deletes the output files generated by this specific stream run.
  • Erase Stream: Removes only the currently selected stream and its logs from the ProgressBoard view.

Log Content

This scrollable area displays the detailed, timestamped logs for the selected agent run.

• Structure: Logs are organized into expandable/collapsible groups (e.g., Initialization, Round 0, Model Operation, Response Cycle) allowing you to focus on specific stages. Click the arrow next to a group name to toggle it.
• Log Levels: Messages are prefixed with levels like INFO, DEBUG, WARN, ERROR to indicate severity. Verbose debug messages (DEBUG) are only shown if texra.logger.verboseOutput is enabled in settings.
• Agent Thinking: Look for entries indicating <scratchpad> content or Thinking: blocks, which show the intermediate reasoning steps of the AI model (especially for CoT agents).
• Errors: Errors are highlighted, often providing clues if something went wrong.

Understanding the log content is key to diagnosing problems and seeing how TeXRA and the AI models process your requests. Refer to the Troubleshooting guide for more tips on using logs.