$npx -y skills add aws/agent-toolkit-for-aws --skill agents-buildUse when adding capabilities to an existing agent project — memory, app integration, VPC, multi-agent, migration, model changes, browser, code interpreter, or resource removal. Triggers on: "add memory", "remember across sessions", "call agent from app", "invoke agent from code",
| 1 | # build |
| 2 | |
| 3 | Add capabilities to your AgentCore agent project. |
| 4 | |
| 5 | ## When to use |
| 6 | |
| 7 | - Adding cross-session memory to your agent |
| 8 | - Calling your deployed agent from a web app, mobile app, or backend service |
| 9 | - Configuring VPC networking for private resources (RDS, internal APIs) |
| 10 | - Building multi-agent systems with orchestrator/specialist patterns |
| 11 | - Migrating an existing Bedrock Agent to AgentCore |
| 12 | - Adding the Browser tool so the agent can navigate websites |
| 13 | - Adding the Code Interpreter so the agent can execute code in a sandbox |
| 14 | - Adding AgentCore Payments so the agent can pay for x402-protected APIs, tools, or content |
| 15 | - Removing resources from your project or tearing down a deployment |
| 16 | |
| 17 | Do NOT use for: |
| 18 | |
| 19 | - Connecting to external tools/APIs via Gateway (OpenAPI specs, Lambda, MCP servers, credentials, policies) → use `agents-connect` |
| 20 | - Scaffolding a new project → use `agents-get-started` |
| 21 | - Deploying → use `agents-deploy` |
| 22 | |
| 23 | ## Input |
| 24 | |
| 25 | `$ARGUMENTS` can be: |
| 26 | |
| 27 | - A capability: "memory", "integrate", "vpc", "multi-agent", "migrate", "browser", "code-interpreter", "payments", "teardown" |
| 28 | - A description of what they want: "remember user preferences", "call from React app", "scrape a website", "run pandas in the agent", "delete my agent", "clean up resources" |
| 29 | - Empty — the skill will determine the workflow from context |
| 30 | |
| 31 | ## Process |
| 32 | |
| 33 | ### Step 0: Verify CLI version |
| 34 | |
| 35 | Run `agentcore --version`. This skill requires v0.9.0 or later. |
| 36 | |
| 37 | If older: "Run `agentcore update` to get the latest version." |
| 38 | |
| 39 | ### Step 1: Read project context |
| 40 | |
| 41 | Read `agentcore/agentcore.json` to understand the current project — framework, existing resources, agent configuration. |
| 42 | |
| 43 | If `agentcore/agentcore.json` is not found: |
| 44 | |
| 45 | 1. **Check if the developer is in the wrong directory.** Look for `agentcore/agentcore.json` in parent directories (up to 3 levels). If found, tell them: "Found an AgentCore project at `<path>`. Are you working in that project?" |
| 46 | 2. **If no project exists anywhere nearby**, ask what capability they wanted to add. Then offer two paths: |
| 47 | - "I can walk you through creating a project first and then adding CAPABILITY — want to do that?" (run the get-started flow inline, then continue with the build workflow) |
| 48 | - "If you already have a project elsewhere, `cd` into it and try again." |
| 49 | |
| 50 | Do not just say "go use agents-get-started" and stop — that loses the developer's context about what they actually wanted to do. |
| 51 | |
| 52 | ### Step 2: Determine the workflow |
| 53 | |
| 54 | **Important disambiguation** — before routing to a build reference, check if the prompt is actually a connect or debug concern: |
| 55 | |
| 56 | - If the phrase mentions external APIs, Lambda functions, OpenAPI specs, gateways, credentials, MCP servers, or policies → this is `agents-connect`, not build |
| 57 | - If the developer says something is broken (wrong answers, errors, tool failures) → this is `agents-debug`, not build |
| 58 | - Build is for **adding new capabilities** to a working project, not fixing broken ones |
| 59 | |
| 60 | Based on the developer's prompt and `$ARGUMENTS`, load the appropriate reference: |
| 61 | |
| 62 | | Developer intent | Reference to load | |
| 63 | |---|---| |
| 64 | | Add memory, remember things, user preferences, cross-session | [`references/memory.md`](references/memory.md) | |
| 65 | | Call agent from app, invoke from code, streaming, SDK client, agent URL, execute shell in session | [`references/integrate.md`](references/integrate.md) | |
| 66 | | VPC, private network, RDS, internal API, subnet, security group | [`references/vpc.md`](references/vpc.md) | |
| 67 | | Multi-agent, orchestrator, specialist, A2A, delegation, agent handoff | [`references/multi-agent.md`](references/multi-agent.md) | |
| 68 | | Custom headers from caller to agent, header allowlist, tenant ID/correlation ID/trace propagation | [`references/request-headers.md`](references/request-h |