This is a meta-layer for MCP itself: it helps Claude discover and recommend other MCP servers based on your project context. Think of it as an app store that analyzes what you're working on and suggests relevant integrations you might want to add. The B2A2H naming (Business to Agent to Human) points to its role as an intermediary that connects available tooling to AI workflows. You'd reach for this when you're unsure which MCP servers exist for your domain or when you want automated suggestions instead of manually browsing catalogs. The auto-discovery piece means it can potentially scan your codebase or project structure and surface relevant integrations without you needing to know they exist.
Public tool metadata for what this MCP can expose to an agent.
search_toolsSearch the global Supernova registry for agent-executable tools by intent.1 paramsSearch the global Supernova registry for agent-executable tools by intent.
intentstringexecute_toolSecurely executes any tool in the registry via the Supernova Proxy.2 paramsSecurely executes any tool in the registry via the Supernova Proxy.
payloadobjecttool_idstringio.github.ericm1018/skillfm-llm-cost-optimizer-openai-anthropic-usage
io.github.mikerawsonnz/llm-orchestration-agent
io.github.mikerawsonnz/authenticated-llm-agent
labforgedev/copilot-memory-mcp
csoai-org/agent-prompt-injection-firewall-mcp
io.github.mikerawsonnz/authenticated-multi-llm-agent