Troubleshooting

If you encounter any issues while using Lumentis, refer to the following troubleshooting guide for help.

Errors During Primary Source Loading

If you're having trouble loading your primary source file or folder, here are a few things to check:

File Type Compatibility: Lumentis currently supports text-based file formats like .txt, .md, .json, .xml, etc. Binary file types like .pdf, .doc, .xls, etc. are not supported.
File Permissions: Ensure that you have the necessary read permissions for the file or folder you're trying to load.
File Size Limits: Lumentis has a token limit for the primary source, which is displayed during the setup process. If your source exceeds this limit, you may need to split it into smaller files.

If you encounter unexpected output or errors during the outline or page generation process, try the following:

Check the Anthropic API Key: Ensure that the Anthropic API key you've provided is valid and has the necessary permissions.
Adjust the AI Model: Try using a different AI model, such as the "Claude 3 Sonnet" or "Claude 3 Haiku" models, which may have better performance for your use case.
Modify the Prompts: If you're an advanced user, you can try modifying the prompts used by Lumentis for outline and page generation. Refer to the Customizing Prompts section for more information.
Check for Partial JSON Responses: Lumentis has built-in handling for partial JSON responses from the language model, but if you're still seeing issues, refer to the Partial JSON Parsing section for more details.

If you're experiencing performance issues, such as slow generation times or high token usage, consider the following:

Optimize the Primary Source: Review the content of your primary source and remove any unnecessary information or formatting. This can help reduce the token count and improve generation times.
Use Worker Threads: Lumentis utilizes worker threads to offload computationally intensive tasks, such as file tree processing and token counting. If you're still experiencing performance issues, refer to the Worker Threads section for more information on how to customize this behavior.
Monitor Token Usage: Lumentis provides cost estimates for the outline and page generation processes. If you're consistently exceeding your budget, review the Token Management section for strategies to optimize token usage.

If you're experiencing issues with integrating Lumentis with other tools or platforms, check the following:

Supported Runners: Lumentis supports various runners (e.g., npm, yarn, pnpm, bun) for setting up the Nextra documentation scaffold. Ensure that the runner you're using is installed and available on your system.
Nextra Compatibility: Lumentis generates Nextra-compatible documentation. If you encounter any issues with the generated pages or the Nextra theme, refer to the Nextra documentation (opens in a new tab) for troubleshooting.
Platform-Specific Issues: Lumentis is designed to be platform-independent, but you may encounter some platform-specific issues. Review the utils.ts (opens in a new tab) file for platform-specific handling and adjust accordingly.

If you're still having trouble, feel free to open an issue on the Lumentis GitHub repository (opens in a new tab) for further assistance.