[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill-aws-agents-build":3,"mdc--ekef18-key":35,"related-org-aws-agents-build":1142,"related-repo-aws-agents-build":1300},{"slug":4,"name":4,"fn":5,"description":6,"org":7,"tags":11,"stars":24,"repoUrl":25,"updatedAt":26,"license":27,"forks":28,"topics":29,"repo":30,"sourceUrl":33,"mdContent":34},"agents-build","add capabilities to existing agent projects","Use 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\", \"auth to call agent\", \"streaming responses\", \"VPC\", \"VPC connectivity\", \"VPC error\", \"can't reach from VPC\", \"multi-agent\", \"A2A\", \"A2A auth\", \"orchestrator not delegating\", \"specialist not called\", \"migrate Bedrock Agent\", \"after import\", \"migration issue\", \"framework for migration\", \"change model\", \"browser tool\", \"code interpreter\", \"delete agent\", \"tear down\", \"agentcore remove\", \"cross-account memory\", \"resource-based policy on memory\", \"pay for x402 content\", \"402 Payment Required\", \"microtransactions\", \"paid API or tool\". Not for connecting to external APIs via Gateway — use agents-connect. Not for scaffolding a new project — use agents-get-started. Not for CLI\u002Fdev server errors — use agents-debug. Strands vs LangGraph in a migration context routes here.\n",{"slug":8,"name":9,"logoUrl":10,"githubOrg":8},"aws","AWS (Amazon)","https:\u002F\u002Fpexgzepcugksgbtrxkhf.supabase.co\u002Fstorage\u002Fv1\u002Fobject\u002Fpublic\u002Forg-logos\u002Faws.png",[12,16,19,22],{"name":13,"slug":14,"type":15},"Automation","automation","tag",{"name":17,"slug":18,"type":15},"Agents","agents",{"name":20,"slug":21,"type":15},"Engineering","engineering",{"name":23,"slug":8,"type":15},"AWS",1822,"https:\u002F\u002Fgithub.com\u002Faws\u002Fagent-toolkit-for-aws","2026-07-12T08:42:53.812877",null,157,[],{"repoUrl":25,"stars":24,"forks":28,"topics":31,"description":32},[],"Official, AWS-supported MCP servers, skills, and plugins to help AI agents build on AWS","https:\u002F\u002Fgithub.com\u002Faws\u002Fagent-toolkit-for-aws\u002Ftree\u002FHEAD\u002Fplugins\u002Faws-agents\u002Fskills\u002Fagents-build","---\nname: agents-build\ndescription: >\n  Use when adding capabilities to an existing agent project — memory,\n  app integration, VPC, multi-agent, migration, model changes, browser,\n  code interpreter, or resource removal. Triggers on: \"add memory\",\n  \"remember across sessions\", \"call agent from app\", \"invoke agent from\n  code\", \"auth to call agent\", \"streaming responses\", \"VPC\", \"VPC\n  connectivity\", \"VPC error\", \"can't reach from VPC\", \"multi-agent\",\n  \"A2A\", \"A2A auth\", \"orchestrator not delegating\", \"specialist not\n  called\", \"migrate Bedrock Agent\", \"after import\", \"migration issue\",\n  \"framework for migration\", \"change model\", \"browser tool\", \"code\n  interpreter\", \"delete agent\", \"tear down\", \"agentcore remove\",\n  \"cross-account memory\", \"resource-based policy on memory\", \"pay for\n  x402 content\", \"402 Payment Required\", \"microtransactions\", \"paid API\n  or tool\".\n  Not for connecting to external APIs via Gateway — use agents-connect.\n  Not for scaffolding a new project — use agents-get-started.\n  Not for CLI\u002Fdev server errors — use agents-debug.\n  Strands vs LangGraph in a migration context routes here.\nallowed-tools: Read Grep Glob Bash\nmetadata:\n  type: skill\n  version: \"1.0.0\"\n  author: aws-agentcore\n  requires-cli: \">=0.9.0\"\n---\n\n# build\n\nAdd capabilities to your AgentCore agent project.\n\n## When to use\n\n- Adding cross-session memory to your agent\n- Calling your deployed agent from a web app, mobile app, or backend service\n- Configuring VPC networking for private resources (RDS, internal APIs)\n- Building multi-agent systems with orchestrator\u002Fspecialist patterns\n- Migrating an existing Bedrock Agent to AgentCore\n- Adding the Browser tool so the agent can navigate websites\n- Adding the Code Interpreter so the agent can execute code in a sandbox\n- Adding AgentCore Payments so the agent can pay for x402-protected APIs, tools, or content\n- Removing resources from your project or tearing down a deployment\n\nDo NOT use for:\n\n- Connecting to external tools\u002FAPIs via Gateway (OpenAPI specs, Lambda, MCP servers, credentials, policies) → use `agents-connect`\n- Scaffolding a new project → use `agents-get-started`\n- Deploying → use `agents-deploy`\n\n## Input\n\n`$ARGUMENTS` can be:\n\n- A capability: \"memory\", \"integrate\", \"vpc\", \"multi-agent\", \"migrate\", \"browser\", \"code-interpreter\", \"payments\", \"teardown\"\n- 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\"\n- Empty — the skill will determine the workflow from context\n\n## Process\n\n### Step 0: Verify CLI version\n\nRun `agentcore --version`. This skill requires v0.9.0 or later.\n\nIf older: \"Run `agentcore update` to get the latest version.\"\n\n### Step 1: Read project context\n\nRead `agentcore\u002Fagentcore.json` to understand the current project — framework, existing resources, agent configuration.\n\nIf `agentcore\u002Fagentcore.json` is not found:\n\n1. **Check if the developer is in the wrong directory.** Look for `agentcore\u002Fagentcore.json` in parent directories (up to 3 levels). If found, tell them: \"Found an AgentCore project at `\u003Cpath>`. Are you working in that project?\"\n2. **If no project exists anywhere nearby**, ask what capability they wanted to add. Then offer two paths:\n   - \"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)\n   - \"If you already have a project elsewhere, `cd` into it and try again.\"\n\nDo not just say \"go use agents-get-started\" and stop — that loses the developer's context about what they actually wanted to do.\n\n### Step 2: Determine the workflow\n\n**Important disambiguation** — before routing to a build reference, check if the prompt is actually a connect or debug concern:\n\n- If the phrase mentions external APIs, Lambda functions, OpenAPI specs, gateways, credentials, MCP servers, or policies → this is `agents-connect`, not build\n- If the developer says something is broken (wrong answers, errors, tool failures) → this is `agents-debug`, not build\n- Build is for **adding new capabilities** to a working project, not fixing broken ones\n\nBased on the developer's prompt and `$ARGUMENTS`, load the appropriate reference:\n\n| Developer intent | Reference to load |\n|---|---|\n| Add memory, remember things, user preferences, cross-session | [`references\u002Fmemory.md`](references\u002Fmemory.md) |\n| Call agent from app, invoke from code, streaming, SDK client, agent URL, execute shell in session | [`references\u002Fintegrate.md`](references\u002Fintegrate.md) |\n| VPC, private network, RDS, internal API, subnet, security group | [`references\u002Fvpc.md`](references\u002Fvpc.md) |\n| Multi-agent, orchestrator, specialist, A2A, delegation, agent handoff | [`references\u002Fmulti-agent.md`](references\u002Fmulti-agent.md) |\n| Custom headers from caller to agent, header allowlist, tenant ID\u002Fcorrelation ID\u002Ftrace propagation | [`references\u002Frequest-headers.md`](references\u002Frequest-headers.md) |\n| Migrate Bedrock Agent, import agent, move to AgentCore | [`references\u002Fmigrate.md`](references\u002Fmigrate.md) |\n| Browser tool, web navigation, form filling, scraping, Nova Act, Playwright, live view | [`references\u002Fbrowser.md`](references\u002Fbrowser.md) |\n| Code Interpreter, execute code, sandbox, run Python\u002FJS\u002FTS, data analysis in agent, pandas | [`references\u002Fcode-interpreter.md`](references\u002Fcode-interpreter.md) |\n| Payments, pay for x402 content, 402 Payment Required, microtransactions, paid API\u002Ftool, payment manager\u002Fconnector | [`references\u002Fpayments.md`](references\u002Fpayments.md) |\n| Delete agent, remove resource, tear down, clean up, destroy, start fresh | [`references\u002Fteardown.md`](references\u002Fteardown.md) |\n| Change model, switch model, use Haiku\u002FSonnet\u002FNova, different model | Inline — see \"Changing the model\" below |\n\nIf the developer asks about the difference between local dev and deployed (e.g., \"why does my memory work after deploy but not locally?\"), load [`references\u002Flocal-vs-deployed.md`](references\u002Flocal-vs-deployed.md) alongside the specific workflow reference.\n\nRead the matching file into context and follow its Process section step by step — do not summarize.\n\nIf the intent is ambiguous, ask the developer which capability they want to add.\n\n### Changing the model\n\nThe model is configured in `app\u002F\u003CAgentName>\u002Fmodel\u002Fload.py` (scaffolded by `agentcore create`). To change it:\n\n1. Open `app\u002F\u003CAgentName>\u002Fmodel\u002Fload.py`\n2. Change the `model_id` parameter in the `BedrockModel()` constructor\n\n```python\n# Default (scaffolded by CLI)\nreturn BedrockModel(model_id=\"global.anthropic.claude-sonnet-4-5-20250929-v1:0\")\n\n# Switch to Haiku for cost savings\nreturn BedrockModel(model_id=\"us.anthropic.claude-3-5-haiku-20241022-v1:0\")\n\n# Switch to Nova Lite\nreturn BedrockModel(model_id=\"amazon.nova-lite-v1:0\")\n```\n\nCross-region inference profile prefixes (`us.`, `eu.`, `apac.`, `global.`) control where inference runs. Use `global.` for maximum throughput, or a geographic prefix for data residency. Not all models support all prefixes — check the Bedrock inference profiles docs.\n\nAfter changing the model:\n\n- Verify the model is enabled in your region: AWS Console → Amazon Bedrock → Model access\n- For cross-region profiles, enable in all destination regions\n- If using `agents-harden`, update the IAM policy to scope to the new model ARN\n- Run `agentcore dev` to test locally, then `agentcore deploy` to update the deployed agent\n\nNo `agentcore.json` change is needed — the model is configured in code, not in the project config.\n\n### Pre-flight: validate any `--name` before generating the CLI command\n\nWhichever reference you load, most end up producing an `agentcore add \u003Cresource> --name \u003Csomething>` command. The CLI fails **late** on invalid names — you'll see the error after walking through prompts, not before running the command. Validate up front:\n\n| Resource | Max chars | Allowed | Starts with |\n|---|---|---|---|\n| Agent (`add agent`) | 48 | alphanumeric + `_` | letter |\n| Memory, gateway, gateway-target, credential, evaluator, online-eval, policy, policy-engine, payment-manager, payment-connector | 48 | alphanumeric + `_` | letter |\n\nCount the characters before constructing the command. If the name is over the limit or contains hyphens, dots, or spaces, push back: \"`\u003Cname>` is N characters \u002F uses `-`, which the CLI rejects. How about `\u003Csuggestion>`?\" Never run the command with an invalid name hoping the CLI message will be clear.\n\nNote: `agentcore create --name` (the project name) has a **stricter 23-char limit** and does not allow underscores. That's covered in `agents-get-started`; if you see the developer re-running create, flag the 23-char limit specifically.\n\n## Output\n\nDepends on the workflow — see the loaded reference for specific outputs.\n\n## Quality criteria\n\n- The correct reference was loaded based on the developer's intent\n- All output follows the loaded reference's quality criteria\n- Cross-references to other skills (agents-connect, agents-deploy) are included where relevant\n",{"data":36,"body":43},{"name":4,"description":6,"allowed-tools":37,"metadata":38},"Read Grep Glob Bash",{"type":39,"version":40,"author":41,"requires-cli":42},"skill","1.0.0","aws-agentcore",">=0.9.0",{"type":44,"children":45},"root",[46,54,60,67,117,122,159,165,176,194,200,207,220,233,239,252,264,325,330,336,346,385,397,639,655,660,665,671,692,726,811,853,858,904,917,931,951,1045,1074,1101,1107,1112,1118,1136],{"type":47,"tag":48,"props":49,"children":51},"element","h1",{"id":50},"build",[52],{"type":53,"value":50},"text",{"type":47,"tag":55,"props":56,"children":57},"p",{},[58],{"type":53,"value":59},"Add capabilities to your AgentCore agent project.",{"type":47,"tag":61,"props":62,"children":64},"h2",{"id":63},"when-to-use",[65],{"type":53,"value":66},"When to use",{"type":47,"tag":68,"props":69,"children":70},"ul",{},[71,77,82,87,92,97,102,107,112],{"type":47,"tag":72,"props":73,"children":74},"li",{},[75],{"type":53,"value":76},"Adding cross-session memory to your agent",{"type":47,"tag":72,"props":78,"children":79},{},[80],{"type":53,"value":81},"Calling your deployed agent from a web app, mobile app, or backend service",{"type":47,"tag":72,"props":83,"children":84},{},[85],{"type":53,"value":86},"Configuring VPC networking for private resources (RDS, internal APIs)",{"type":47,"tag":72,"props":88,"children":89},{},[90],{"type":53,"value":91},"Building multi-agent systems with orchestrator\u002Fspecialist patterns",{"type":47,"tag":72,"props":93,"children":94},{},[95],{"type":53,"value":96},"Migrating an existing Bedrock Agent to AgentCore",{"type":47,"tag":72,"props":98,"children":99},{},[100],{"type":53,"value":101},"Adding the Browser tool so the agent can navigate websites",{"type":47,"tag":72,"props":103,"children":104},{},[105],{"type":53,"value":106},"Adding the Code Interpreter so the agent can execute code in a sandbox",{"type":47,"tag":72,"props":108,"children":109},{},[110],{"type":53,"value":111},"Adding AgentCore Payments so the agent can pay for x402-protected APIs, tools, or content",{"type":47,"tag":72,"props":113,"children":114},{},[115],{"type":53,"value":116},"Removing resources from your project or tearing down a deployment",{"type":47,"tag":55,"props":118,"children":119},{},[120],{"type":53,"value":121},"Do NOT use for:",{"type":47,"tag":68,"props":123,"children":124},{},[125,137,148],{"type":47,"tag":72,"props":126,"children":127},{},[128,130],{"type":53,"value":129},"Connecting to external tools\u002FAPIs via Gateway (OpenAPI specs, Lambda, MCP servers, credentials, policies) → use ",{"type":47,"tag":131,"props":132,"children":134},"code",{"className":133},[],[135],{"type":53,"value":136},"agents-connect",{"type":47,"tag":72,"props":138,"children":139},{},[140,142],{"type":53,"value":141},"Scaffolding a new project → use ",{"type":47,"tag":131,"props":143,"children":145},{"className":144},[],[146],{"type":53,"value":147},"agents-get-started",{"type":47,"tag":72,"props":149,"children":150},{},[151,153],{"type":53,"value":152},"Deploying → use ",{"type":47,"tag":131,"props":154,"children":156},{"className":155},[],[157],{"type":53,"value":158},"agents-deploy",{"type":47,"tag":61,"props":160,"children":162},{"id":161},"input",[163],{"type":53,"value":164},"Input",{"type":47,"tag":55,"props":166,"children":167},{},[168,174],{"type":47,"tag":131,"props":169,"children":171},{"className":170},[],[172],{"type":53,"value":173},"$ARGUMENTS",{"type":53,"value":175}," can be:",{"type":47,"tag":68,"props":177,"children":178},{},[179,184,189],{"type":47,"tag":72,"props":180,"children":181},{},[182],{"type":53,"value":183},"A capability: \"memory\", \"integrate\", \"vpc\", \"multi-agent\", \"migrate\", \"browser\", \"code-interpreter\", \"payments\", \"teardown\"",{"type":47,"tag":72,"props":185,"children":186},{},[187],{"type":53,"value":188},"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\"",{"type":47,"tag":72,"props":190,"children":191},{},[192],{"type":53,"value":193},"Empty — the skill will determine the workflow from context",{"type":47,"tag":61,"props":195,"children":197},{"id":196},"process",[198],{"type":53,"value":199},"Process",{"type":47,"tag":201,"props":202,"children":204},"h3",{"id":203},"step-0-verify-cli-version",[205],{"type":53,"value":206},"Step 0: Verify CLI version",{"type":47,"tag":55,"props":208,"children":209},{},[210,212,218],{"type":53,"value":211},"Run ",{"type":47,"tag":131,"props":213,"children":215},{"className":214},[],[216],{"type":53,"value":217},"agentcore --version",{"type":53,"value":219},". This skill requires v0.9.0 or later.",{"type":47,"tag":55,"props":221,"children":222},{},[223,225,231],{"type":53,"value":224},"If older: \"Run ",{"type":47,"tag":131,"props":226,"children":228},{"className":227},[],[229],{"type":53,"value":230},"agentcore update",{"type":53,"value":232}," to get the latest version.\"",{"type":47,"tag":201,"props":234,"children":236},{"id":235},"step-1-read-project-context",[237],{"type":53,"value":238},"Step 1: Read project context",{"type":47,"tag":55,"props":240,"children":241},{},[242,244,250],{"type":53,"value":243},"Read ",{"type":47,"tag":131,"props":245,"children":247},{"className":246},[],[248],{"type":53,"value":249},"agentcore\u002Fagentcore.json",{"type":53,"value":251}," to understand the current project — framework, existing resources, agent configuration.",{"type":47,"tag":55,"props":253,"children":254},{},[255,257,262],{"type":53,"value":256},"If ",{"type":47,"tag":131,"props":258,"children":260},{"className":259},[],[261],{"type":53,"value":249},{"type":53,"value":263}," is not found:",{"type":47,"tag":265,"props":266,"children":267},"ol",{},[268,294],{"type":47,"tag":72,"props":269,"children":270},{},[271,277,279,284,286,292],{"type":47,"tag":272,"props":273,"children":274},"strong",{},[275],{"type":53,"value":276},"Check if the developer is in the wrong directory.",{"type":53,"value":278}," Look for ",{"type":47,"tag":131,"props":280,"children":282},{"className":281},[],[283],{"type":53,"value":249},{"type":53,"value":285}," in parent directories (up to 3 levels). If found, tell them: \"Found an AgentCore project at ",{"type":47,"tag":131,"props":287,"children":289},{"className":288},[],[290],{"type":53,"value":291},"\u003Cpath>",{"type":53,"value":293},". Are you working in that project?\"",{"type":47,"tag":72,"props":295,"children":296},{},[297,302,304],{"type":47,"tag":272,"props":298,"children":299},{},[300],{"type":53,"value":301},"If no project exists anywhere nearby",{"type":53,"value":303},", ask what capability they wanted to add. Then offer two paths:\n",{"type":47,"tag":68,"props":305,"children":306},{},[307,312],{"type":47,"tag":72,"props":308,"children":309},{},[310],{"type":53,"value":311},"\"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)",{"type":47,"tag":72,"props":313,"children":314},{},[315,317,323],{"type":53,"value":316},"\"If you already have a project elsewhere, ",{"type":47,"tag":131,"props":318,"children":320},{"className":319},[],[321],{"type":53,"value":322},"cd",{"type":53,"value":324}," into it and try again.\"",{"type":47,"tag":55,"props":326,"children":327},{},[328],{"type":53,"value":329},"Do not just say \"go use agents-get-started\" and stop — that loses the developer's context about what they actually wanted to do.",{"type":47,"tag":201,"props":331,"children":333},{"id":332},"step-2-determine-the-workflow",[334],{"type":53,"value":335},"Step 2: Determine the workflow",{"type":47,"tag":55,"props":337,"children":338},{},[339,344],{"type":47,"tag":272,"props":340,"children":341},{},[342],{"type":53,"value":343},"Important disambiguation",{"type":53,"value":345}," — before routing to a build reference, check if the prompt is actually a connect or debug concern:",{"type":47,"tag":68,"props":347,"children":348},{},[349,361,373],{"type":47,"tag":72,"props":350,"children":351},{},[352,354,359],{"type":53,"value":353},"If the phrase mentions external APIs, Lambda functions, OpenAPI specs, gateways, credentials, MCP servers, or policies → this is ",{"type":47,"tag":131,"props":355,"children":357},{"className":356},[],[358],{"type":53,"value":136},{"type":53,"value":360},", not build",{"type":47,"tag":72,"props":362,"children":363},{},[364,366,372],{"type":53,"value":365},"If the developer says something is broken (wrong answers, errors, tool failures) → this is ",{"type":47,"tag":131,"props":367,"children":369},{"className":368},[],[370],{"type":53,"value":371},"agents-debug",{"type":53,"value":360},{"type":47,"tag":72,"props":374,"children":375},{},[376,378,383],{"type":53,"value":377},"Build is for ",{"type":47,"tag":272,"props":379,"children":380},{},[381],{"type":53,"value":382},"adding new capabilities",{"type":53,"value":384}," to a working project, not fixing broken ones",{"type":47,"tag":55,"props":386,"children":387},{},[388,390,395],{"type":53,"value":389},"Based on the developer's prompt and ",{"type":47,"tag":131,"props":391,"children":393},{"className":392},[],[394],{"type":53,"value":173},{"type":53,"value":396},", load the appropriate reference:",{"type":47,"tag":398,"props":399,"children":400},"table",{},[401,420],{"type":47,"tag":402,"props":403,"children":404},"thead",{},[405],{"type":47,"tag":406,"props":407,"children":408},"tr",{},[409,415],{"type":47,"tag":410,"props":411,"children":412},"th",{},[413],{"type":53,"value":414},"Developer intent",{"type":47,"tag":410,"props":416,"children":417},{},[418],{"type":53,"value":419},"Reference to load",{"type":47,"tag":421,"props":422,"children":423},"tbody",{},[424,446,466,486,506,526,546,566,586,606,626],{"type":47,"tag":406,"props":425,"children":426},{},[427,433],{"type":47,"tag":428,"props":429,"children":430},"td",{},[431],{"type":53,"value":432},"Add memory, remember things, user preferences, cross-session",{"type":47,"tag":428,"props":434,"children":435},{},[436],{"type":47,"tag":437,"props":438,"children":440},"a",{"href":439},"references\u002Fmemory.md",[441],{"type":47,"tag":131,"props":442,"children":444},{"className":443},[],[445],{"type":53,"value":439},{"type":47,"tag":406,"props":447,"children":448},{},[449,454],{"type":47,"tag":428,"props":450,"children":451},{},[452],{"type":53,"value":453},"Call agent from app, invoke from code, streaming, SDK client, agent URL, execute shell in session",{"type":47,"tag":428,"props":455,"children":456},{},[457],{"type":47,"tag":437,"props":458,"children":460},{"href":459},"references\u002Fintegrate.md",[461],{"type":47,"tag":131,"props":462,"children":464},{"className":463},[],[465],{"type":53,"value":459},{"type":47,"tag":406,"props":467,"children":468},{},[469,474],{"type":47,"tag":428,"props":470,"children":471},{},[472],{"type":53,"value":473},"VPC, private network, RDS, internal API, subnet, security group",{"type":47,"tag":428,"props":475,"children":476},{},[477],{"type":47,"tag":437,"props":478,"children":480},{"href":479},"references\u002Fvpc.md",[481],{"type":47,"tag":131,"props":482,"children":484},{"className":483},[],[485],{"type":53,"value":479},{"type":47,"tag":406,"props":487,"children":488},{},[489,494],{"type":47,"tag":428,"props":490,"children":491},{},[492],{"type":53,"value":493},"Multi-agent, orchestrator, specialist, A2A, delegation, agent handoff",{"type":47,"tag":428,"props":495,"children":496},{},[497],{"type":47,"tag":437,"props":498,"children":500},{"href":499},"references\u002Fmulti-agent.md",[501],{"type":47,"tag":131,"props":502,"children":504},{"className":503},[],[505],{"type":53,"value":499},{"type":47,"tag":406,"props":507,"children":508},{},[509,514],{"type":47,"tag":428,"props":510,"children":511},{},[512],{"type":53,"value":513},"Custom headers from caller to agent, header allowlist, tenant ID\u002Fcorrelation ID\u002Ftrace propagation",{"type":47,"tag":428,"props":515,"children":516},{},[517],{"type":47,"tag":437,"props":518,"children":520},{"href":519},"references\u002Frequest-headers.md",[521],{"type":47,"tag":131,"props":522,"children":524},{"className":523},[],[525],{"type":53,"value":519},{"type":47,"tag":406,"props":527,"children":528},{},[529,534],{"type":47,"tag":428,"props":530,"children":531},{},[532],{"type":53,"value":533},"Migrate Bedrock Agent, import agent, move to AgentCore",{"type":47,"tag":428,"props":535,"children":536},{},[537],{"type":47,"tag":437,"props":538,"children":540},{"href":539},"references\u002Fmigrate.md",[541],{"type":47,"tag":131,"props":542,"children":544},{"className":543},[],[545],{"type":53,"value":539},{"type":47,"tag":406,"props":547,"children":548},{},[549,554],{"type":47,"tag":428,"props":550,"children":551},{},[552],{"type":53,"value":553},"Browser tool, web navigation, form filling, scraping, Nova Act, Playwright, live view",{"type":47,"tag":428,"props":555,"children":556},{},[557],{"type":47,"tag":437,"props":558,"children":560},{"href":559},"references\u002Fbrowser.md",[561],{"type":47,"tag":131,"props":562,"children":564},{"className":563},[],[565],{"type":53,"value":559},{"type":47,"tag":406,"props":567,"children":568},{},[569,574],{"type":47,"tag":428,"props":570,"children":571},{},[572],{"type":53,"value":573},"Code Interpreter, execute code, sandbox, run Python\u002FJS\u002FTS, data analysis in agent, pandas",{"type":47,"tag":428,"props":575,"children":576},{},[577],{"type":47,"tag":437,"props":578,"children":580},{"href":579},"references\u002Fcode-interpreter.md",[581],{"type":47,"tag":131,"props":582,"children":584},{"className":583},[],[585],{"type":53,"value":579},{"type":47,"tag":406,"props":587,"children":588},{},[589,594],{"type":47,"tag":428,"props":590,"children":591},{},[592],{"type":53,"value":593},"Payments, pay for x402 content, 402 Payment Required, microtransactions, paid API\u002Ftool, payment manager\u002Fconnector",{"type":47,"tag":428,"props":595,"children":596},{},[597],{"type":47,"tag":437,"props":598,"children":600},{"href":599},"references\u002Fpayments.md",[601],{"type":47,"tag":131,"props":602,"children":604},{"className":603},[],[605],{"type":53,"value":599},{"type":47,"tag":406,"props":607,"children":608},{},[609,614],{"type":47,"tag":428,"props":610,"children":611},{},[612],{"type":53,"value":613},"Delete agent, remove resource, tear down, clean up, destroy, start fresh",{"type":47,"tag":428,"props":615,"children":616},{},[617],{"type":47,"tag":437,"props":618,"children":620},{"href":619},"references\u002Fteardown.md",[621],{"type":47,"tag":131,"props":622,"children":624},{"className":623},[],[625],{"type":53,"value":619},{"type":47,"tag":406,"props":627,"children":628},{},[629,634],{"type":47,"tag":428,"props":630,"children":631},{},[632],{"type":53,"value":633},"Change model, switch model, use Haiku\u002FSonnet\u002FNova, different model",{"type":47,"tag":428,"props":635,"children":636},{},[637],{"type":53,"value":638},"Inline — see \"Changing the model\" below",{"type":47,"tag":55,"props":640,"children":641},{},[642,644,653],{"type":53,"value":643},"If the developer asks about the difference between local dev and deployed (e.g., \"why does my memory work after deploy but not locally?\"), load ",{"type":47,"tag":437,"props":645,"children":647},{"href":646},"references\u002Flocal-vs-deployed.md",[648],{"type":47,"tag":131,"props":649,"children":651},{"className":650},[],[652],{"type":53,"value":646},{"type":53,"value":654}," alongside the specific workflow reference.",{"type":47,"tag":55,"props":656,"children":657},{},[658],{"type":53,"value":659},"Read the matching file into context and follow its Process section step by step — do not summarize.",{"type":47,"tag":55,"props":661,"children":662},{},[663],{"type":53,"value":664},"If the intent is ambiguous, ask the developer which capability they want to add.",{"type":47,"tag":201,"props":666,"children":668},{"id":667},"changing-the-model",[669],{"type":53,"value":670},"Changing the model",{"type":47,"tag":55,"props":672,"children":673},{},[674,676,682,684,690],{"type":53,"value":675},"The model is configured in ",{"type":47,"tag":131,"props":677,"children":679},{"className":678},[],[680],{"type":53,"value":681},"app\u002F\u003CAgentName>\u002Fmodel\u002Fload.py",{"type":53,"value":683}," (scaffolded by ",{"type":47,"tag":131,"props":685,"children":687},{"className":686},[],[688],{"type":53,"value":689},"agentcore create",{"type":53,"value":691},"). To change it:",{"type":47,"tag":265,"props":693,"children":694},{},[695,705],{"type":47,"tag":72,"props":696,"children":697},{},[698,700],{"type":53,"value":699},"Open ",{"type":47,"tag":131,"props":701,"children":703},{"className":702},[],[704],{"type":53,"value":681},{"type":47,"tag":72,"props":706,"children":707},{},[708,710,716,718,724],{"type":53,"value":709},"Change the ",{"type":47,"tag":131,"props":711,"children":713},{"className":712},[],[714],{"type":53,"value":715},"model_id",{"type":53,"value":717}," parameter in the ",{"type":47,"tag":131,"props":719,"children":721},{"className":720},[],[722],{"type":53,"value":723},"BedrockModel()",{"type":53,"value":725}," constructor",{"type":47,"tag":727,"props":728,"children":733},"pre",{"className":729,"code":730,"language":731,"meta":732,"style":732},"language-python shiki shiki-themes material-theme-lighter material-theme material-theme-palenight","# Default (scaffolded by CLI)\nreturn BedrockModel(model_id=\"global.anthropic.claude-sonnet-4-5-20250929-v1:0\")\n\n# Switch to Haiku for cost savings\nreturn BedrockModel(model_id=\"us.anthropic.claude-3-5-haiku-20241022-v1:0\")\n\n# Switch to Nova Lite\nreturn BedrockModel(model_id=\"amazon.nova-lite-v1:0\")\n","python","",[734],{"type":47,"tag":131,"props":735,"children":736},{"__ignoreMap":732},[737,748,757,767,776,785,793,802],{"type":47,"tag":738,"props":739,"children":742},"span",{"class":740,"line":741},"line",1,[743],{"type":47,"tag":738,"props":744,"children":745},{},[746],{"type":53,"value":747},"# Default (scaffolded by CLI)\n",{"type":47,"tag":738,"props":749,"children":751},{"class":740,"line":750},2,[752],{"type":47,"tag":738,"props":753,"children":754},{},[755],{"type":53,"value":756},"return BedrockModel(model_id=\"global.anthropic.claude-sonnet-4-5-20250929-v1:0\")\n",{"type":47,"tag":738,"props":758,"children":760},{"class":740,"line":759},3,[761],{"type":47,"tag":738,"props":762,"children":764},{"emptyLinePlaceholder":763},true,[765],{"type":53,"value":766},"\n",{"type":47,"tag":738,"props":768,"children":770},{"class":740,"line":769},4,[771],{"type":47,"tag":738,"props":772,"children":773},{},[774],{"type":53,"value":775},"# Switch to Haiku for cost savings\n",{"type":47,"tag":738,"props":777,"children":779},{"class":740,"line":778},5,[780],{"type":47,"tag":738,"props":781,"children":782},{},[783],{"type":53,"value":784},"return BedrockModel(model_id=\"us.anthropic.claude-3-5-haiku-20241022-v1:0\")\n",{"type":47,"tag":738,"props":786,"children":788},{"class":740,"line":787},6,[789],{"type":47,"tag":738,"props":790,"children":791},{"emptyLinePlaceholder":763},[792],{"type":53,"value":766},{"type":47,"tag":738,"props":794,"children":796},{"class":740,"line":795},7,[797],{"type":47,"tag":738,"props":798,"children":799},{},[800],{"type":53,"value":801},"# Switch to Nova Lite\n",{"type":47,"tag":738,"props":803,"children":805},{"class":740,"line":804},8,[806],{"type":47,"tag":738,"props":807,"children":808},{},[809],{"type":53,"value":810},"return BedrockModel(model_id=\"amazon.nova-lite-v1:0\")\n",{"type":47,"tag":55,"props":812,"children":813},{},[814,816,822,824,830,831,837,838,844,846,851],{"type":53,"value":815},"Cross-region inference profile prefixes (",{"type":47,"tag":131,"props":817,"children":819},{"className":818},[],[820],{"type":53,"value":821},"us.",{"type":53,"value":823},", ",{"type":47,"tag":131,"props":825,"children":827},{"className":826},[],[828],{"type":53,"value":829},"eu.",{"type":53,"value":823},{"type":47,"tag":131,"props":832,"children":834},{"className":833},[],[835],{"type":53,"value":836},"apac.",{"type":53,"value":823},{"type":47,"tag":131,"props":839,"children":841},{"className":840},[],[842],{"type":53,"value":843},"global.",{"type":53,"value":845},") control where inference runs. Use ",{"type":47,"tag":131,"props":847,"children":849},{"className":848},[],[850],{"type":53,"value":843},{"type":53,"value":852}," for maximum throughput, or a geographic prefix for data residency. Not all models support all prefixes — check the Bedrock inference profiles docs.",{"type":47,"tag":55,"props":854,"children":855},{},[856],{"type":53,"value":857},"After changing the model:",{"type":47,"tag":68,"props":859,"children":860},{},[861,866,871,884],{"type":47,"tag":72,"props":862,"children":863},{},[864],{"type":53,"value":865},"Verify the model is enabled in your region: AWS Console → Amazon Bedrock → Model access",{"type":47,"tag":72,"props":867,"children":868},{},[869],{"type":53,"value":870},"For cross-region profiles, enable in all destination regions",{"type":47,"tag":72,"props":872,"children":873},{},[874,876,882],{"type":53,"value":875},"If using ",{"type":47,"tag":131,"props":877,"children":879},{"className":878},[],[880],{"type":53,"value":881},"agents-harden",{"type":53,"value":883},", update the IAM policy to scope to the new model ARN",{"type":47,"tag":72,"props":885,"children":886},{},[887,888,894,896,902],{"type":53,"value":211},{"type":47,"tag":131,"props":889,"children":891},{"className":890},[],[892],{"type":53,"value":893},"agentcore dev",{"type":53,"value":895}," to test locally, then ",{"type":47,"tag":131,"props":897,"children":899},{"className":898},[],[900],{"type":53,"value":901},"agentcore deploy",{"type":53,"value":903}," to update the deployed agent",{"type":47,"tag":55,"props":905,"children":906},{},[907,909,915],{"type":53,"value":908},"No ",{"type":47,"tag":131,"props":910,"children":912},{"className":911},[],[913],{"type":53,"value":914},"agentcore.json",{"type":53,"value":916}," change is needed — the model is configured in code, not in the project config.",{"type":47,"tag":201,"props":918,"children":920},{"id":919},"pre-flight-validate-any-name-before-generating-the-cli-command",[921,923,929],{"type":53,"value":922},"Pre-flight: validate any ",{"type":47,"tag":131,"props":924,"children":926},{"className":925},[],[927],{"type":53,"value":928},"--name",{"type":53,"value":930}," before generating the CLI command",{"type":47,"tag":55,"props":932,"children":933},{},[934,936,942,944,949],{"type":53,"value":935},"Whichever reference you load, most end up producing an ",{"type":47,"tag":131,"props":937,"children":939},{"className":938},[],[940],{"type":53,"value":941},"agentcore add \u003Cresource> --name \u003Csomething>",{"type":53,"value":943}," command. The CLI fails ",{"type":47,"tag":272,"props":945,"children":946},{},[947],{"type":53,"value":948},"late",{"type":53,"value":950}," on invalid names — you'll see the error after walking through prompts, not before running the command. Validate up front:",{"type":47,"tag":398,"props":952,"children":953},{},[954,980],{"type":47,"tag":402,"props":955,"children":956},{},[957],{"type":47,"tag":406,"props":958,"children":959},{},[960,965,970,975],{"type":47,"tag":410,"props":961,"children":962},{},[963],{"type":53,"value":964},"Resource",{"type":47,"tag":410,"props":966,"children":967},{},[968],{"type":53,"value":969},"Max chars",{"type":47,"tag":410,"props":971,"children":972},{},[973],{"type":53,"value":974},"Allowed",{"type":47,"tag":410,"props":976,"children":977},{},[978],{"type":53,"value":979},"Starts with",{"type":47,"tag":421,"props":981,"children":982},{},[983,1020],{"type":47,"tag":406,"props":984,"children":985},{},[986,999,1004,1015],{"type":47,"tag":428,"props":987,"children":988},{},[989,991,997],{"type":53,"value":990},"Agent (",{"type":47,"tag":131,"props":992,"children":994},{"className":993},[],[995],{"type":53,"value":996},"add agent",{"type":53,"value":998},")",{"type":47,"tag":428,"props":1000,"children":1001},{},[1002],{"type":53,"value":1003},"48",{"type":47,"tag":428,"props":1005,"children":1006},{},[1007,1009],{"type":53,"value":1008},"alphanumeric + ",{"type":47,"tag":131,"props":1010,"children":1012},{"className":1011},[],[1013],{"type":53,"value":1014},"_",{"type":47,"tag":428,"props":1016,"children":1017},{},[1018],{"type":53,"value":1019},"letter",{"type":47,"tag":406,"props":1021,"children":1022},{},[1023,1028,1032,1041],{"type":47,"tag":428,"props":1024,"children":1025},{},[1026],{"type":53,"value":1027},"Memory, gateway, gateway-target, credential, evaluator, online-eval, policy, policy-engine, payment-manager, payment-connector",{"type":47,"tag":428,"props":1029,"children":1030},{},[1031],{"type":53,"value":1003},{"type":47,"tag":428,"props":1033,"children":1034},{},[1035,1036],{"type":53,"value":1008},{"type":47,"tag":131,"props":1037,"children":1039},{"className":1038},[],[1040],{"type":53,"value":1014},{"type":47,"tag":428,"props":1042,"children":1043},{},[1044],{"type":53,"value":1019},{"type":47,"tag":55,"props":1046,"children":1047},{},[1048,1050,1056,1058,1064,1066,1072],{"type":53,"value":1049},"Count the characters before constructing the command. If the name is over the limit or contains hyphens, dots, or spaces, push back: \"",{"type":47,"tag":131,"props":1051,"children":1053},{"className":1052},[],[1054],{"type":53,"value":1055},"\u003Cname>",{"type":53,"value":1057}," is N characters \u002F uses ",{"type":47,"tag":131,"props":1059,"children":1061},{"className":1060},[],[1062],{"type":53,"value":1063},"-",{"type":53,"value":1065},", which the CLI rejects. How about ",{"type":47,"tag":131,"props":1067,"children":1069},{"className":1068},[],[1070],{"type":53,"value":1071},"\u003Csuggestion>",{"type":53,"value":1073},"?\" Never run the command with an invalid name hoping the CLI message will be clear.",{"type":47,"tag":55,"props":1075,"children":1076},{},[1077,1079,1085,1087,1092,1094,1099],{"type":53,"value":1078},"Note: ",{"type":47,"tag":131,"props":1080,"children":1082},{"className":1081},[],[1083],{"type":53,"value":1084},"agentcore create --name",{"type":53,"value":1086}," (the project name) has a ",{"type":47,"tag":272,"props":1088,"children":1089},{},[1090],{"type":53,"value":1091},"stricter 23-char limit",{"type":53,"value":1093}," and does not allow underscores. That's covered in ",{"type":47,"tag":131,"props":1095,"children":1097},{"className":1096},[],[1098],{"type":53,"value":147},{"type":53,"value":1100},"; if you see the developer re-running create, flag the 23-char limit specifically.",{"type":47,"tag":61,"props":1102,"children":1104},{"id":1103},"output",[1105],{"type":53,"value":1106},"Output",{"type":47,"tag":55,"props":1108,"children":1109},{},[1110],{"type":53,"value":1111},"Depends on the workflow — see the loaded reference for specific outputs.",{"type":47,"tag":61,"props":1113,"children":1115},{"id":1114},"quality-criteria",[1116],{"type":53,"value":1117},"Quality criteria",{"type":47,"tag":68,"props":1119,"children":1120},{},[1121,1126,1131],{"type":47,"tag":72,"props":1122,"children":1123},{},[1124],{"type":53,"value":1125},"The correct reference was loaded based on the developer's intent",{"type":47,"tag":72,"props":1127,"children":1128},{},[1129],{"type":53,"value":1130},"All output follows the loaded reference's quality criteria",{"type":47,"tag":72,"props":1132,"children":1133},{},[1134],{"type":53,"value":1135},"Cross-references to other skills (agents-connect, agents-deploy) are included where relevant",{"type":47,"tag":1137,"props":1138,"children":1139},"style",{},[1140],{"type":53,"value":1141},"html .light .shiki span {color: var(--shiki-light);background: var(--shiki-light-bg);font-style: var(--shiki-light-font-style);font-weight: var(--shiki-light-font-weight);text-decoration: var(--shiki-light-text-decoration);}html.light .shiki span {color: var(--shiki-light);background: var(--shiki-light-bg);font-style: var(--shiki-light-font-style);font-weight: var(--shiki-light-font-weight);text-decoration: var(--shiki-light-text-decoration);}html .default .shiki span {color: var(--shiki-default);background: var(--shiki-default-bg);font-style: var(--shiki-default-font-style);font-weight: var(--shiki-default-font-weight);text-decoration: var(--shiki-default-text-decoration);}html .shiki span {color: var(--shiki-default);background: var(--shiki-default-bg);font-style: var(--shiki-default-font-style);font-weight: var(--shiki-default-font-weight);text-decoration: var(--shiki-default-text-decoration);}html .dark .shiki span {color: var(--shiki-dark);background: var(--shiki-dark-bg);font-style: var(--shiki-dark-font-style);font-weight: var(--shiki-dark-font-weight);text-decoration: var(--shiki-dark-text-decoration);}html.dark .shiki span {color: var(--shiki-dark);background: var(--shiki-dark-bg);font-style: var(--shiki-dark-font-style);font-weight: var(--shiki-dark-font-weight);text-decoration: var(--shiki-dark-text-decoration);}",{"items":1143,"total":1299},[1144,1151,1165,1179,1193,1202,1216,1232,1249,1262,1274,1289],{"slug":4,"name":4,"fn":5,"description":6,"org":1145,"tags":1146,"stars":24,"repoUrl":25,"updatedAt":26},{"slug":8,"name":9,"logoUrl":10,"githubOrg":8},[1147,1148,1149,1150],{"name":17,"slug":18,"type":15},{"name":13,"slug":14,"type":15},{"name":23,"slug":8,"type":15},{"name":20,"slug":21,"type":15},{"slug":136,"name":136,"fn":1152,"description":1153,"org":1154,"tags":1155,"stars":24,"repoUrl":25,"updatedAt":1164},"connect agents to external services","Use when connecting your agent to external APIs, tools, or services via Gateway, or restricting tool access with Cedar policies. Handles gateway setup, target types, outbound auth (OAuth, API key, IAM), credentials, and Cedar policy authoring. Triggers on: \"connect to API\", \"add gateway\", \"connect to MCP server\", \"Lambda tools\", \"OpenAPI\", \"gateway target\", \"Cedar policy\", \"restrict tools\", \"policy engine\", \"gateway auth error\", \"store API key\", \"outbound credential\", \"env var API key\", \"API key None after deploy\", \"credential not available after deploy\", \"should this be a gateway target\", \"give my agent tools\", \"add tools to agent\". Not for inbound auth (who can call your agent) — use agents-harden. Not for debugging agent behavior — use agents-debug. Not for VPC networking errors (agent can't reach APIs due to VPC) — use agents-build. Not for creating or hosting a new MCP server project — use agents-get-started.\n",{"slug":8,"name":9,"logoUrl":10,"githubOrg":8},[1156,1157,1160,1163],{"name":17,"slug":18,"type":15},{"name":1158,"slug":1159,"type":15},"API Development","api-development",{"name":1161,"slug":1162,"type":15},"Authentication","authentication",{"name":23,"slug":8,"type":15},"2026-07-16T06:00:38.866147",{"slug":371,"name":371,"fn":1166,"description":1167,"org":1168,"tags":1169,"stars":24,"repoUrl":25,"updatedAt":1178},"debug agent and environment issues","Use when your agent or environment is broken — wrong answers, errors, timeouts, tool failures, or CLI issues. Reads traces and logs to diagnose root causes. Also checks prerequisites when the CLI itself isn't working. Triggers on: \"agent not working\", \"wrong answer\", \"agent error\", \"tool call failing\", \"debug agent\", \"check logs\", \"read traces\", \"broken\", \"500 error\", \"424 error\", \"model access denied\", \"command not found\", \"stuck in DELETING\", \"maxVms exceeded\", \"cold start diagnosis\", \"cold start slow\", \"agentcore create error\", \"create failed\", \"exit code 7\", \"connection refused local dev\". Not for deploy failures — use agents-deploy. Not for performance tuning without errors — use agents-optimize. Not for VPC configuration — use agents-build. Not for observability setup or missing logs — use agents-optimize.\n",{"slug":8,"name":9,"logoUrl":10,"githubOrg":8},[1170,1171,1172,1175],{"name":17,"slug":18,"type":15},{"name":23,"slug":8,"type":15},{"name":1173,"slug":1174,"type":15},"Debugging","debugging",{"name":1176,"slug":1177,"type":15},"Observability","observability","2026-07-16T06:00:44.679093",{"slug":158,"name":158,"fn":1180,"description":1181,"org":1182,"tags":1183,"stars":24,"repoUrl":25,"updatedAt":1192},"deploy AI agents to AWS","Use when deploying your agent to AWS, or when a deploy has failed. Handles pre-flight validation, CDK\u002FIAM\u002Fquota error diagnosis, version management, rollback, and canary deployments. Triggers on: \"deploy my agent\", \"agentcore deploy\", \"deploy failed\", \"CDK error\", \"rollback\", \"canary deploy\", \"pin version\", \"redeploy\", \"deploy stuck\". Not for production hardening — use agents-harden. Not for adding capabilities before deploy — use agents-build or agents-connect. Not for VPC configuration errors — use agents-build.\n",{"slug":8,"name":9,"logoUrl":10,"githubOrg":8},[1184,1185,1186,1189],{"name":17,"slug":18,"type":15},{"name":23,"slug":8,"type":15},{"name":1187,"slug":1188,"type":15},"CI\u002FCD","ci-cd",{"name":1190,"slug":1191,"type":15},"Deployment","deployment","2026-07-12T08:42:55.059577",{"slug":147,"name":147,"fn":1194,"description":1195,"org":1196,"tags":1197,"stars":24,"repoUrl":25,"updatedAt":1201},"scaffold and deploy new agent projects","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.\n",{"slug":8,"name":9,"logoUrl":10,"githubOrg":8},[1198,1199,1200],{"name":17,"slug":18,"type":15},{"name":23,"slug":8,"type":15},{"name":1190,"slug":1191,"type":15},"2026-07-12T08:42:51.963247",{"slug":881,"name":881,"fn":1203,"description":1204,"org":1205,"tags":1206,"stars":24,"repoUrl":25,"updatedAt":1215},"harden agents for production","Use when preparing your agent for production — IAM scoping, inbound auth (JWT, SigV4), secrets management, cold start optimization, session lifecycle, rate limiting, input validation, and quota guidance. Triggers on: \"production checklist\", \"harden agent\", \"production ready\", \"secure agent\", \"inbound auth\", \"going live\", \"cold start optimization\", \"session lifecycle\", \"StopRuntimeSession\", \"quota\", \"throttling\", \"maxVms\", \"rate limit\", \"security audit of outbound API calls\", \"gateway target audit for production\", \"restrict who can call\", \"lock down endpoint\", \"only our app can call\". Not for Cedar tool-restriction policies — use agents-connect. Not for quality measurement — use agents-optimize. Not for outbound credential storage or API key wiring — use agents-connect. Not for A2A agent-to-agent auth — use agents-build. Cold start observation and diagnosis (not optimization) routes to agents-debug.\n",{"slug":8,"name":9,"logoUrl":10,"githubOrg":8},[1207,1208,1209,1212],{"name":17,"slug":18,"type":15},{"name":23,"slug":8,"type":15},{"name":1210,"slug":1211,"type":15},"Best Practices","best-practices",{"name":1213,"slug":1214,"type":15},"Security","security","2026-07-16T06:00:42.174705",{"slug":1217,"name":1217,"fn":1218,"description":1219,"org":1220,"tags":1221,"stars":24,"repoUrl":25,"updatedAt":1231},"agents-optimize","optimize agent quality and performance","Use when measuring or improving agent quality and performance — set up evaluators, online monitoring, CI\u002FCD quality gates, observability, or cost optimization. Triggers on: \"evaluate my agent\", \"add evaluator\", \"measure quality\", \"quality gate\", \"run evals\", \"agent too slow\", \"why is it slow\", \"reduce latency\", \"set up observability\", \"CloudWatch dashboard\", \"how much does my agent cost\", \"cost optimization\", \"logs not showing up\", \"logs missing\", \"spans not found\", \"eval failing\", \"eval error\", \"dev traces\", \"local traces\", \"agentcore dev traces\", \"traces to CloudWatch\". Not for debugging errors or crashes — use agents-debug. Slow but correct routes here; broken routes to debug.\n",{"slug":8,"name":9,"logoUrl":10,"githubOrg":8},[1222,1223,1224,1227,1228],{"name":17,"slug":18,"type":15},{"name":23,"slug":8,"type":15},{"name":1225,"slug":1226,"type":15},"Evals","evals",{"name":1176,"slug":1177,"type":15},{"name":1229,"slug":1230,"type":15},"Performance","performance","2026-07-12T08:42:56.488105",{"slug":1233,"name":1233,"fn":1234,"description":1235,"org":1236,"tags":1237,"stars":24,"repoUrl":25,"updatedAt":1248},"amazon-aurora-mysql","manage Amazon Aurora MySQL clusters","Amazon Aurora MySQL — creates, modifies, and advises on Aurora MySQL clusters specifically (MySQL-compatible engine, Aurora serverless, parallel query). Trigger for Aurora MySQL cluster operations, ACU sizing, I\u002FO-Optimized storage, commitment pricing, or MySQL upgrade planning. Aurora MySQL uses full (VPC-based) configuration — express configuration is PostgreSQL-only. For Aurora PostgreSQL, use amazon-aurora-postgresql instead. Contains safety guardrails and response templates that override defaults.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":8},[1238,1239,1242,1245],{"name":23,"slug":8,"type":15},{"name":1240,"slug":1241,"type":15},"Database","database",{"name":1243,"slug":1244,"type":15},"MySQL","mysql",{"name":1246,"slug":1247,"type":15},"Serverless","serverless","2026-07-12T08:43:13.27939",{"slug":1250,"name":1250,"fn":1251,"description":1252,"org":1253,"tags":1254,"stars":24,"repoUrl":25,"updatedAt":1261},"amazon-aurora-postgresql","configure Amazon Aurora PostgreSQL clusters","Amazon Aurora PostgreSQL — creates, modifies, and advises on Aurora PostgreSQL clusters specifically (PostgreSQL-compatible engine, Aurora serverless, express configuration, pgvector, Babelfish). Trigger for Aurora PostgreSQL cluster operations, express-configuration quick-start, ACU sizing, I\u002FO-Optimized storage, commitment pricing, or PostgreSQL upgrade planning. For Aurora MySQL, use amazon-aurora-mysql instead. Contains safety guardrails, express-first routing, and response templates that override defaults.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":8},[1255,1256,1257,1260],{"name":23,"slug":8,"type":15},{"name":1240,"slug":1241,"type":15},{"name":1258,"slug":1259,"type":15},"PostgreSQL","postgresql",{"name":1246,"slug":1247,"type":15},"2026-07-16T06:00:34.789624",{"slug":1263,"name":1263,"fn":1264,"description":1265,"org":1266,"tags":1267,"stars":24,"repoUrl":25,"updatedAt":1273},"amazon-bedrock","build generative AI apps with Amazon Bedrock","Builds generative AI applications on Amazon Bedrock. Covers model invocation (Converse API, InvokeModel), RAG with Knowledge Bases, Bedrock Agents, Guardrails, and AgentCore. Use when invoking models, setting up Knowledge Bases, creating agents, applying guardrails, deploying to AgentCore, migrating\u002Fporting\u002Fconverting a Bedrock Agent (including inline agents) to an AgentCore Harness, troubleshooting Bedrock errors (ThrottlingException, AccessDeniedException), or choosing models (Claude, Llama, Nova, Titan). ALSO USE for prompt caching setup and debugging, quota health checks and throttling diagnosis, cost attribution and tracking, migrating between Claude model generations (4.5 to 4.6 to 4.7), chunking strategies, API selection (Converse vs InvokeModel), guardrail capabilities, and model selection. Also covers AgentCore Payments setup (x402, microtransactions, Payment Manager, Connector, Instrument, Coinbase CDP, Stripe Privy, 402 Payment Required, pay for content, paid endpoint, agent payments). NOT for custom model training, Rekognition, or Comprehend.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":8},[1268,1269,1270],{"name":17,"slug":18,"type":15},{"name":23,"slug":8,"type":15},{"name":1271,"slug":1272,"type":15},"LLM","llm","2026-07-25T05:30:35.20899",{"slug":1275,"name":1275,"fn":1276,"description":1277,"org":1278,"tags":1279,"stars":24,"repoUrl":25,"updatedAt":1288},"amazon-documentdb","manage Amazon DocumentDB clusters","Manages Amazon DocumentDB end-to-end — serverless-on-8.0 cluster setup, TLS\u002FVPC\u002Fdriver config, flexible-schema and vector-search data modeling, MongoDB compatibility assessment, DMS-based migration, slow-query diagnosis, major version upgrades (4.0→5.0→8.0), Well-Architected reviews (41-check wa_review.py), cost estimation, and security hardening. Retrieve for every DocumentDB question and when the user asks to set up or migrate MongoDB to AWS — DocumentDB is AWS's MongoDB-compatible managed database. Triggers: JSON document store, document database, MongoDB on AWS, Nested fields, Lambda cannot connect, TLS handshake, VPC port 27017, IAM auth, Secrets Manager, encryption at rest, $graphLookup, flexible schema, COLLSCAN, compound index, DMS migration, CDC cutover, $vectorSearch, RAG, Global Clusters, DR replication, cost sizing, audit, health check, production-readiness.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":8},[1280,1281,1282,1285],{"name":23,"slug":8,"type":15},{"name":1240,"slug":1241,"type":15},{"name":1283,"slug":1284,"type":15},"MongoDB","mongodb",{"name":1286,"slug":1287,"type":15},"NoSQL","nosql","2026-07-12T08:43:00.455878",{"slug":1290,"name":1290,"fn":1291,"description":1292,"org":1293,"tags":1294,"stars":24,"repoUrl":25,"updatedAt":1298},"amazon-dynamodb","design and debug DynamoDB data layers","Designs, reviews, and debugs DynamoDB data layers from design axioms — enumerates access patterns, chooses partition\u002Fsort keys and GSIs, decides single-table vs. multi-table, configures Streams, Global Tables, TTL, and zero-ETL integrations to OpenSearch\u002FRedshift\u002FSageMaker Lakehouse, and produces a defensible data-layer design with a monthly cost estimate and optional live validation. Applies whenever a user is designing, reviewing, or refactoring anything backed by DynamoDB — schemas, access patterns, GSIs, single- vs. multi-table choices, Streams consumers, transactional outboxes, Global Tables, zero-ETL pipelines — even when they don't say \"axioms\" or \"design review.\" Also applies when debugging hot partitions, throttling, unbounded Scans, LWW conflicts, or surprise bills on DynamoDB workloads.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":8},[1295,1296,1297],{"name":23,"slug":8,"type":15},{"name":1240,"slug":1241,"type":15},{"name":1286,"slug":1287,"type":15},"2026-07-16T06:00:37.690386",115,{"items":1301,"total":1351},[1302,1309,1316,1323,1330,1336,1343],{"slug":4,"name":4,"fn":5,"description":6,"org":1303,"tags":1304,"stars":24,"repoUrl":25,"updatedAt":26},{"slug":8,"name":9,"logoUrl":10,"githubOrg":8},[1305,1306,1307,1308],{"name":17,"slug":18,"type":15},{"name":13,"slug":14,"type":15},{"name":23,"slug":8,"type":15},{"name":20,"slug":21,"type":15},{"slug":136,"name":136,"fn":1152,"description":1153,"org":1310,"tags":1311,"stars":24,"repoUrl":25,"updatedAt":1164},{"slug":8,"name":9,"logoUrl":10,"githubOrg":8},[1312,1313,1314,1315],{"name":17,"slug":18,"type":15},{"name":1158,"slug":1159,"type":15},{"name":1161,"slug":1162,"type":15},{"name":23,"slug":8,"type":15},{"slug":371,"name":371,"fn":1166,"description":1167,"org":1317,"tags":1318,"stars":24,"repoUrl":25,"updatedAt":1178},{"slug":8,"name":9,"logoUrl":10,"githubOrg":8},[1319,1320,1321,1322],{"name":17,"slug":18,"type":15},{"name":23,"slug":8,"type":15},{"name":1173,"slug":1174,"type":15},{"name":1176,"slug":1177,"type":15},{"slug":158,"name":158,"fn":1180,"description":1181,"org":1324,"tags":1325,"stars":24,"repoUrl":25,"updatedAt":1192},{"slug":8,"name":9,"logoUrl":10,"githubOrg":8},[1326,1327,1328,1329],{"name":17,"slug":18,"type":15},{"name":23,"slug":8,"type":15},{"name":1187,"slug":1188,"type":15},{"name":1190,"slug":1191,"type":15},{"slug":147,"name":147,"fn":1194,"description":1195,"org":1331,"tags":1332,"stars":24,"repoUrl":25,"updatedAt":1201},{"slug":8,"name":9,"logoUrl":10,"githubOrg":8},[1333,1334,1335],{"name":17,"slug":18,"type":15},{"name":23,"slug":8,"type":15},{"name":1190,"slug":1191,"type":15},{"slug":881,"name":881,"fn":1203,"description":1204,"org":1337,"tags":1338,"stars":24,"repoUrl":25,"updatedAt":1215},{"slug":8,"name":9,"logoUrl":10,"githubOrg":8},[1339,1340,1341,1342],{"name":17,"slug":18,"type":15},{"name":23,"slug":8,"type":15},{"name":1210,"slug":1211,"type":15},{"name":1213,"slug":1214,"type":15},{"slug":1217,"name":1217,"fn":1218,"description":1219,"org":1344,"tags":1345,"stars":24,"repoUrl":25,"updatedAt":1231},{"slug":8,"name":9,"logoUrl":10,"githubOrg":8},[1346,1347,1348,1349,1350],{"name":17,"slug":18,"type":15},{"name":23,"slug":8,"type":15},{"name":1225,"slug":1226,"type":15},{"name":1176,"slug":1177,"type":15},{"name":1229,"slug":1230,"type":15},114]