ac360-capability-router

Helps choose or confirm a benchmark subject for portfolio comparison. Use when asked what benchmark fits, to compare a portfolio to a model or index, to set a benchmark, or to use an advisor-saved portfolio as the benchmark. Skip when a benchmark is already selected and the advisor is asking for the analysis itself.

Builds, imports, modifies or clones portfolios after the advisor confirms the holdings and write action. Use when asked to build from tickers or weights, create a portfolio from a statement, start from a model or saved portfolio, create a comparison portfolio, or explore a specific what-if variation. Skip when the portfolio already exists and the advisor is only asking for review or analysis.

Reviews returned portfolio opportunity checks, fund-health details, Funds to Explore, and approved follow-up links. Use when asked to show opportunities, explain why a fund is flagged, open fund details, show fund health, list clear checks, find Funds to Explore, or identify the next path from an opportunity. Skip when the advisor asks for a broad portfolio review before narrowing to opportunities.

Reviews a saved portfolio, household or model with Advisor Center analytics in three sections rendered inline in the chat — portfolio snapshot, stress scenarios, then opportunities, fund health and Funds to Explore — closing with follow-ups earned by what it found. Use when asked to review a portfolio, run a portfolio review, analyse holdings, pull up a client's account, run the numbers on a portfolio, look at concentration or downside exposure, find issues or opportunities in a portfolio, check whether a client's funds are healthy, or prepare for a client review meeting. Skip this skill when the advisor already named one specific analysis — "what's the risk on this", "run the stress scenarios", "how does this compare to the model" — and call that tool directly instead of running the three-section pass.

Explains modeled performance, projected wealth, methodology, advisory-fee treatment, and unsupported performance or projection requests. Use when asked to show modeled performance, project wealth, explain projection assumptions, check whether fees are included, or handle an unavailable horizon or configuration. Skip when the advisor asks for a broad portfolio review or only needs benchmark setup.

Configure this repo for the engineering skills: set up its issue tracker, triage label vocabulary, and domain doc layout. Run once before first use of the other engineering skills.

Teach the user a new skill or concept, within this workspace.

Ask which skill or flow fits your situation. A router over the skills in this repo.

Plan a huge chunk of work (more than one agent session can hold) as a shared map of decision tickets on your issue tracker, and resolve them one at a time until the way to the destination is clear.

Hand the current conversation off to a fresh background agent that picks up the work immediately.

Create new skills, modify and improve existing skills, and measure skill performance. Use when users want to create a skill from scratch, edit, or optimize an existing skill, run evals to test a skill, benchmark skill performance with variance analysis, or optimize a skill's description for better triggering accuracy.

Use when a developer wants to create a new agent project or get started with AgentCore. Handles framework selection, project scaffolding, first deploy, and first invocation. Triggers on: "build an agent", "create an agent", "get started", "new project", "agentcore create", "which framework", "Strands vs LangGraph", "hello world agent", "first agent", "create MCP server", "host MCP server", "agentcore dev", "dev server", "what port", "local development". Not for adding capabilities to existing projects — use agents-build or agents-connect. Strands vs LangGraph in a migration context routes to agents-build, not here. Connecting to an existing MCP server routes to agents-connect, not here.

Connects AI agents to remote Windows desktop applications on Amazon WorkSpaces Applications (AppStream 2.0) through the managed Agent Access MCP server, and guides reliable desktop automation. Covers connecting an agent to the MCP endpoint (SigV4, streaming URL, and Active Directory SAML/Domain Join), BLOCKING vs POLLING connect modes, the computer-use tools (screenshot, click, type, key, scroll), screenshot-budget and action-batching discipline, MCP tool forwarding (forwarded___ tools), session lifecycle and expire-on-delete, and troubleshooting connection errors. Use when building or debugging an agent that drives a remote Windows desktop or GUI application via WorkSpaces Applications / AppStream — including "dcv session not ready", "client_disconnected", 400 signing-region, POLLING/connection_status, SAML assertion, or forwarded tool questions. Not for Amazon WorkSpaces Personal/Core virtual desktops or general AppStream fleet administration unrelated to agent access.