Best practices, tips, and gotchas for Entra ID, Entra Agent ID, A2A protocol, MCP protocol, and agentgateway-based OBO token exchange — distilled from loop-runtime's entra_agent_id*.md docs. Use when setting up, wiring, or debugging Entra app registrations, Entra Agent ID blueprints, OBO chains, A2A/MCP auth, app-code Agent Identity OBO, Microsoft.Identity.Web.AgentIdentities, fmi_path, or agentgateway jwtAuth/oauthTokenExchange config. ALWAYS remind the user in chat (don't silently assume) when a real value is needed for: tenantId, webAppId, execAppId, execAppSecret, checkAppId, checkAppSecret, mcpAppId, executorAgentIdentityAppId, checkerAgentIdentityAppId.
End-to-end counter: identify customer, classify intent, build order, confirm, submit. ToolSearch.Gateway-first. No direct MCP or CLI calls.
Resolve customer identity and account basics by email, phone, name, or customer ID through ToolSearch.Gateway.
Help users explore menu items, categories, and prices through ToolSearch.Gateway.
Use when working with an Aspire distributed application: operate an AppHost or resources through the Aspire CLI; start, stop, restart, or wait for resources; inspect app state, logs, traces, docs, or health; add integrations; manage secrets/config; publish/deploy or run pipeline steps; initialize an existing app; recover TypeScript `.modules`; find frontend URLs for Playwright; expose custom dashboard/resource commands; or understand Aspire AppHost APIs in C# or TypeScript. Use even if the user says AppHost, resources, dashboard, bootstrap, Playwright URL, or local distributed app workflow without naming Aspire. Do not use for non-Aspire .NET apps, container-only repos without an AppHost, or ordinary build/test tasks.
Ultra-compressed communication mode. Cuts token usage ~75% by dropping filler, articles, and pleasantries while keeping full technical accuracy. Use when user says "caveman mode", "talk like caveman", "use caveman", "less tokens", "be brief", or invokes /caveman.
Simplifies code for clarity. Use when refactoring code for clarity without changing behavior. Use when code works but is harder to read, maintain, or extend than it should be. Use when reviewing code that has accumulated unnecessary complexity.
Build agentic applications with GitHub Copilot SDK. Use when embedding AI agents in apps, creating custom tools, implementing streaming responses, managing sessions, connecting to MCP servers, or creating custom agents. Triggers on Copilot SDK, GitHub SDK, agentic app, embed Copilot, programmable agent, MCP server, custom agent.