BytePlus LLM Application Firewall · emploidai Marketplace
emploidai Marketplace
Add-onsAppletsPlugins
Search tools, teams, and capabilitiesPublish
MarketplacePluginsBytePlus LLM Application Firewall
Plugin
Limited listing

BytePlus LLM Application Firewall

by byteplus · v0.0.1

The LLM Application Firewall provides security protection services for LLM inference services, ensuring that the input and output content of the models is secure, available, and trustworthy. The product is embedded into the business processes of AI large model services, monitoring the input and output content in real time to protect model operations from OWASP LLM Top 10 attacks. It offers various protective functions, including computing power consumption protection, prompt attack detection, model abuse behavior analysis, and sensitive data risk identification.

108 installsUpdated Jan 29, 2026
Publisher information is incomplete

This community listing does not yet include every recommended support, privacy, pricing, and permission disclosure. Review the available package permissions before installing.

Capabilities

Tools

Available inside your emploidai workspace after installation.

Data sources

Available inside your emploidai workspace after installation.

Category

tool

Version

0.0.1byteplus

Requirements

Maximum memory 256MB

Pricing

Not disclosed by publisher

Security & access

Review before installing

CompatibleRequires emploidai 1.0.0+

Permissions

  • Uses tool capability
  • Requires encrypted tool credentials

Dependencies

No additional dependencies

Resources

Privacy policy
emploidai Marketplace

Discover capabilities. Review access. Install inside your workspace.

DocumentationSecuritySupportPrivacyTerms

The BytePlus LLM Application Firewall is a comprehensive security solution designed to provide robust protection for applications leveraging large model inference. This product effectively defends against:

  • Computational DDoS attacks
  • Prompt injection attacks
  • Sensitive data leakage
  • Model abuse Furthermore, to cater to diverse business scenarios, this product features intelligent topic control. This allows for adaptive adjustment of protection policies, ensuring the secure and stable operation of your large model applications.

Getting Started

  1. Log in to the BytePlus LLM Application Firewall Console.
  2. If you have not activated it yet, follow the page instructions to activate a BytePlus LLM Application Firewall instance.
  3. Integrate your assets using the SDK. For detailed instructions, please refer to Configure SDK Integration Parameters. Take note of the instance region and App ID, as they will be required when calling the tools.
  4. Create and obtain the Access Key (AK) and Secret Key (SK) for your BytePlus account. For detailed instructions, please refer to Obtain AccessKey and SecretKey.
  5. Configure the AK and SK within the plugin.

"Check for content compliance" Tool Description

Call this tool to determine if the model's input or output content is compliant. To understand the specific meaning of each field, please refer to API Documentation.

  • Request parameters:
    • Region: Select the region where your LLM Application Firewall instance is located (e.g., ap-southeast-1).
    • AppID: Enter the AppID of your SDK asset, which can be found on the Asset Management page of the LLM Application Firewall.
    • Role: Select the source of the content to be checked.
      • User: User prompts
      • Assistant: Model responses
      • System: Default prompts
    • Content: The specific content that needs to be checked. Supports text, URLs, or Base64 (must include a header like 'data:application/msword;base64,' for images/documents).
    • Contents: Content list for mixed inspection (JSON array format, e.g., [{"content": "text content", "content_type": "text"}, {"content": "data:image/png;base64,...", "content_type": "image"}]). Available content_type values: text, image, file. This parameter will be used when the content type is selected as "Mixed".
    • ContentType: Select the type of content to be checked. Supports Text, Image, File, Mixed. Note: Document detection only supports the following 6 formats: PDF, DOCX, DOC, XLSX, XLS, TXT.
  • Responses: The tool will write the results to a JSON output value. To see how to reference this, you can run the tool once and examine the specific output.

"Generate safe alternative response" Tool Description

This tool generates safe, alternative responses. To use it, you must configure the action for your protection rules to "Safe Alternative Response" in the console. For a detailed explanation of the fields, please refer to API Documentation.

  1. Obtain the msg_id directly from the output results of the "Check for content compliance" tool.
  2. Configure the following parameters. The alternative response will be written to the text field.
    • Region: Select the region where your LLM Application Firewall instance is located (e.g., ap-southeast-1).
    • MessageID: Enter the msg_id obtained from the output of the inspection tool.