Adversa AI coding agent security platform
Named by 3 kept items in this workspace.
Search every workspace for Adversa AI coding agent security platform. It matches the name as written, so it can find other things that share it.
The items that name it 3
Links every model call, tool call, and endpoint action into chains, downgrades an agent's trust the moment it touches untrusted content, and blocks dangerous chains before completion; used to reproduce and successfully stop the CCI chain against an instrumented coding agent.
Commercial runtime security layer (Agent Security Gateway, optionally with an Endpoint Security Sensor) that watches agent traffic, correlates untrusted input, file/secret access and commands against the agent's stated reasoning, and ships 100+ attack-anchored policies to block or pause risky agenti
Links every model call, tool call, and endpoint action into chains, downgrades an agent's trust the moment it touches untrusted content, and blocks dangerous chains before completion; demonstrated stopping the Grok-style Cryptographic Context Injection chain against an instrumented coding agent.
All technologies in this workspace.