hmr-cookbook scored 100/100 on AgentSeal (https://agentseal.org). We put it through AI semantic analysis, sandbox extraction, and live adversarial probing. One tool (echo), minimal attack surface, nothing concerning.
One small note: the echo tool has no description. It's a pretty self-explanatory name, but a short description helps agents reason about when to use it and what to expect back.
Full report: https://agentseal.org/mcp/hmr-cookbook
[](https://agentseal.org/mcp/hmr-cookbook)
If anything looks off, let us know.
hmr-cookbook scored 100/100 on AgentSeal (https://agentseal.org). We put it through AI semantic analysis, sandbox extraction, and live adversarial probing. One tool (echo), minimal attack surface, nothing concerning.
One small note: the echo tool has no description. It's a pretty self-explanatory name, but a short description helps agents reason about when to use it and what to expect back.
Full report: https://agentseal.org/mcp/hmr-cookbook
If anything looks off, let us know.