← All work

RiFT

Agentic security systems

Agent-assisted black-box web/API security research with MCP tooling, remote bridges and isolated execution.

FieldSecurity / AI
AffiliationRiFT Research · SENTIA Lab
StatusActive engineering

Overview

RiFT is an agent-assisted black-box web/API security research framework. Work centers on integrating MCP-based tools, remote bridges, scoped execution and isolated sandboxes.

The project is under active engineering. Features and integration boundaries continue to evolve and should not be interpreted as a stable security product.

Areas of focus

  • Agent orchestration and tool selection
  • Sandbox boundaries and execution safety
  • Web and backend security assessment
  • MCP and plugin-based tool integration

Technology & concepts

PythonSecurity testingSandboxingMCP

Source & references

View on GitHub