THREAT OPS › Threat News › BloodHound MCP, One Year Later: What I Learned About MCPs, Models, and Context
BloodHound MCP, One Year Later: What I Learned About MCPs, Models, and Context
<p class="wp-block-paragraph"><em><strong>TL:DR: </strong>The first version of BloodHound MCP proved that an LLM could converse with BloodHound. The current version drove home the lesson that MCP design is context design. The most useful changes were smaller but more flexible tools, better error recovery, domain-specific resources, and a prompt that guides the model instead of trying to teach ever
MITRE ATT&CK techniques
- KerberoastingT1558.003
- System PromptAML.T0069.002