superagent-ai/superagent

Superagent protects your AI applications against prompt injections, data leaks, and harmful outputs. Embed safety directly into your app and prove compliance to your customers.

What it solves

Superagent provides a security layer for AI agents to prevent them from being manipulated or leaking sensitive data. It addresses risks like prompt injections, malicious tool calls, and the accidental exposure of personally identifiable information (PII) or secrets.

How it works

The SDK integrates into AI applications via TypeScript or Python to provide four primary security functions:

  • Guard: Monitors runtime inputs to detect and block prompt injections and unsafe instructions.
  • Redact: Automatically identifies and removes PII, PHI, and secrets from text.
  • Scan: Analyzes code repositories for AI-specific threats like repo poisoning.
  • Test: Executes red-teaming scenarios to find vulnerabilities in production agents.

Who it’s for

Developers building AI agents who need to ensure their applications are safe, compliant, and resilient against adversarial attacks.

Highlights

  • Model Agnostic: Compatible with major providers including OpenAI, Anthropic, Google, Groq, and Bedrock.
  • Low Latency: Optimized for runtime use with open-weight models that can run on local infrastructure.
  • Multiple Integrations: Available as TypeScript and Python SDKs, a CLI, and an MCP server for Claude.

Related

  • Project
    arcjet/arcjet-js

    A runtime security platform for AI applications that detects prompt injection, redacts sensitive data, and manages bot abuse and rate limiting.

  • Project
    AISecurityLab/hackagent

    A security red-teaming toolkit for AI agents that automates the discovery of vulnerabilities like prompt injection and jailbreaking using research-backed attack techniques.

  • Project
    cisco-ai-defense/skill-scanner

    A security scanner for AI Agent Skills that detects prompt injection, data exfiltration, and malicious code using static analysis, behavioral dataflow, and LLM-based semantic analysis.

  • Dispatch
    OpenAI Designing AI Agents to Resist Prompt Injection

    OpenAI is shifting its defense strategy against prompt injection by treating it as a social engineering problem, focusing on constraining the impact of successful manipulations rather than relying solely on input filtering.

  • Project
    asamassekou10/ship-safe

    An independent security agent for AI-written software that finds, investigates, and verifies vulnerabilities in code, LLM prompts, and agent configurations.