Browsing Tag
Prompt injection
3 posts
MCP prompt injection: limit the agent’s blast radius
MCP prompt injection: cut what an agent can do with narrow tools, a minimal token and human approval, then test the guardrails and trace every call.
SkillSpector agent skills security scanner: install and read
SkillSpector agent skills security scanner: install NVIDIA's tool from the README, read the risk score and plug it into your CI pipeline.
MCP tool poisoning: detecting it with traces and logs
MCP tool poisoning: fingerprint the tool catalogue, read OpenTelemetry spans and alert in Dynatrace to catch line jumping, shadowing and rug pulls.