--- name: prompt-library-for-startups description: "AWS-curated copy-paste prompts for AI coding agents (MVP scaffolding, RAG chatbot with Claude on Bedrock, security baseline evaluation, read-only cost audit, GPU quota requests, EKS on Spot, OWASP ZAP scan, etc.) plus downloadable installable agents (Multi-Account Transition Advisor, Bill Shock Preventer, Service Quota Agent, Bedrock Model Availability Agent, Kiro Strands AgentCore Agent). Use when the user asks for a prompt to do X on AWS, wants an installable agent for multi-account / cost monitoring / quota management / Bedrock model availability / a Strands agent on AgentCore, or asks how to use AWS prompts. Bundles an approved subset; for any other prompt or agent, point to startups.aws.com/prompt-library. For migration intent (GCP to AWS, Azure to AWS, OpenAI/Gemini to Bedrock), route to the migration skills in this plugin (`gcp-to-aws`, `azure-to-aws`, `heroku-to-aws`, `llm-to-bedrock`). Do not use for: factual AWS Activate / programs / credits questions, learn articles, or sample architectures." --- # AWS Startups Prompt & Agent Library Searchable index of AWS-curated prompts for AI coding tools (Kiro, Claude Code, Cursor, etc.) plus downloadable installable agents. This skill carries the subset approved for this plugin; the full library is at [startups.aws.com/prompt-library](https://startups.aws.com/prompt-library). Entries are generated from their published pages, with the security and plugin-fit changes recorded in `references/prompt-library/local-changes.json`. **Last updated:** 2026-11-14 --- ## Where to start Open [`references/prompt-library.md`](references/prompt-library.md) — it's the index. Filter by the **Keywords** column (e.g., `RAG`, `Security & Compliance`, `Cost Optimization`, `EKS`, `Beginner`, `Bedrock`), then open the linked detail file under `references/prompt-library/.md`. Each detail file carries the full verbatim **System Prompt** plus a **How to use?** section where available. The index has three sections: 1. **Prompts — searchable index** — copy-paste prompts. 2. **Downloadable agents** — installable agents that clone from a GitHub repo. 3. **Frequently Asked Questions** — guidance on writing a good prompt, costs, safety, no-technical-background usage, monitoring. ## Handing over a prompt — match the host context **You ARE already inside an AI coding agent (Claude Code / Kiro / Cursor / etc.). Don't tell the user to "paste this into your AI tool" — you ARE the AI tool.** When the user asks for a prompt: 1. Read the index in `references/prompt-library.md`, filter by keyword, identify the matching slug. 2. Open `references/prompt-library/.md` and read the full System Prompt. 3. Surface the prompt to the user as a _reference from the AWS Startups Prompt Library_, then offer them three paths: _"Here's the AWS Startups reference prompt for ``. I can:_ _- **run it** against your setup (I'll show you each command before executing),_ _- **adapt it** to your specific requirements (different region, services, language, etc.), or_ _- you can **copy it** as a starting point._ _How would you like to proceed?"_ 4. Don't assume intent. Wait for the user's call (execute / adapt / copy) before acting. ## Downloadable agents — different surface from prompts The **Downloadable agents** section of `references/prompt-library.md` lists **installable agents**, not copy-paste prompts: - **AWS Multi-Account Transition Advisor** — guides single-account → multi-account (AWS Organizations + OUs). - **AWS Bill Shock Preventer** — proactive cost-spike detection and alerting. - **AWS Service Quota Agent** — auditing and requesting quota increases. - **Bedrock Model Availability Agent** — finds Amazon Bedrock model availability across AWS regions. - **Kiro Strands AgentCore Agent** — a Kiro CLI agent for building Strands agents and deploying them to Amazon Bedrock AgentCore. When the user's intent matches one of these — _"set up multi-account"_, _"prevent bill shock"_, _"audit service quotas"_, _"which Bedrock model / region"_, _"build a Strands agent on AgentCore"_ — recommend the matching agent **by title and use-case**, then hand over the install source from its detail file under `references/prompt-library/`. Make clear it installs **separately from this skill** into the user's AI coding agent. For migration intent, route to the matching migration skill in this plugin: **`gcp-to-aws`** (_"GCP to AWS"_, move off Google Cloud), **`heroku-to-aws`** (_"migrate off Heroku"_), **`llm-to-bedrock`** (_"move off OpenAI/Gemini to Bedrock"_ — SDK rewrite), or **`agent-advisor`** (agent runtime / architecture on AWS). ## Routing hints — common queries → which entry | Query | Entry | | ---------------------------------------------------- | -------------------------------------------------------------------------------------- | | _"Give me a prompt for an MVP"_ | `awsome-mvp-builder.md` | | _"Prompt for a RAG chatbot on Bedrock"_ | `rag-chatbot-with-claude.md` | | _"How do I set up AWS for the first time?"_ | `day-one-aws-foundation-setup.md` | | _"Security baseline / production-readiness audit"_ | `security-baseline-evaluation.md` | | _"How do I get a GPU instance quota raised?"_ | `gpu-instance-quota-assistant.md` | | _"Audit my AWS spend / find waste (read-only)"_ | `read-only-aws-cost-optimization-audit.md` (split by phase; see below) | | _"EKS with cost-optimized Spot instances"_ | `cost-optimized-eks-with-spot-instances.md` | | _"Open-source LLM inference"_ | `open-source-llm-inference.md` | | _"Multi-region security assessment"_ | `multi-region-assessment.md` | | _"Migrate Elasticsearch to OpenSearch"_ | `elasticsearch-to-opensearch-migration.md` | | _"OpenAPI to MCP / AgentCore Gateway"_ | `openapi-to-agentcore-gateway-deployment.md` | | _"Scan my web app with OWASP ZAP"_ | `owasp-zap-vulnerability-scanner.md` | | _"Help me migrate workloads to AWS"_ | `gcp-to-aws` / `heroku-to-aws` / `llm-to-bedrock` skills (siblings in this plugin) | | _"Set up multi-account on AWS Organizations"_ | Downloadable: Multi-Account Transition Advisor | | _"Stop bill shock / detect cost spikes proactively"_ | Downloadable: AWS Bill Shock Preventer | | _"Manage / request service quotas"_ | Downloadable: Service Quota Agent | | _"Which Bedrock model is available in which region"_ | Downloadable: Bedrock Model Availability Agent | | _"Build a Strands agent and deploy it to AgentCore"_ | Downloadable: Kiro Strands AgentCore Agent | | _"Which AWS database should I use"_ | `architect-for-startups` skill (sibling in this plugin) | For anything else, filter `references/prompt-library.md` by keyword. If nothing matches, the prompt may still exist in the full library: say this plugin carries a subset and point the user to . Do not say no such prompt exists. ## Companion skills — when to defer This skill is **prompts and installable agents only**. Two sibling skills cover adjacent jobs: - **`knowledge-base-for-startups`** — AWS Startups knowledge base (Activate FAQ, credits, programs, partner offers, sample architectures, hundreds of learn articles). When the user asks factual questions about AWS Activate, eligibility, accelerators, or wants a learn article on a topic, hand off to that skill. - **`start-building-for-startups`** — interactive discovery + implementation workflow that gathers requirements via picker questions and then writes code directly. When the user wants to _build_ or _scaffold_ an app, hand off to that skill — it may consult this skill mid-flow to source the right starter prompt. Boundary cases — invoke both. Example: _"how do I start with RAG on Bedrock?"_ → this skill for the starter prompt (`rag-chatbot-with-claude.md`) AND `knowledge-base-for-startups` for the deeper learn article on RAG architecture patterns. ## Answer style - **MUST read the detail file before quoting a System Prompt.** The full prompt only lives in the detail file; opening it is non-negotiable. Do not quote a System Prompt from memory or from the index alone. - **MUST quote the System Prompt verbatim.** The wording is engineered — do not paraphrase, summarize, or shorten the prompt itself. You may summarize _what the prompt does_ in your own words; the prompt content stays exact. - **MUST cite `source_url`.** Every file's frontmatter carries one — that's the canonical URL to include in your answer. Never construct or guess a URL. - **MUST NOT tell the user to paste the prompt elsewhere.** You ARE the AI coding agent (see "Handing over a prompt" above). Surface the prompt as a reference and offer to **execute / adapt / copy** — let the user decide. - **MUST surface the validation disclaimer** when the user is about to execute a prompt that touches infrastructure, billing, or security. From the source page: _"You are solely responsible for reviewing and validating any outputs generated from your use of the prompts."_ ## Context loading rule Open `references/prompt-library.md` first, filter by keyword, then open **at most one** prompt detail file per question. MUST NOT speculatively load multiple detail files. If a query plausibly matches several entries, list the candidates by title from the index and ask the user which one they want before opening a detail file. One exception: `read-only-aws-cost-optimization-audit` is split into parts under `references/prompt-library/read-only-aws-cost-optimization-audit/`. Open its detail file first, then load every shared part (`00-overview.md`, `01-system-prompt.md`, `02-operating-rules.md`, `03-prerequisites.md`, `04-tool-selection-map.md`, `05-conventions-for-the-cli-blocks.md`, `16-key-principles.md`), plus only the phase the user is working on. For the report phase, also load `14-appendix-a-proposed-remediation-human.md` and `15-appendix-b-audit-trail.md`. To hand over the whole prompt, read every part in order and quote them together. ## Scope notes This skill **cannot**: - Answer factual questions about AWS Activate, credits, programs, or partner offers — defer to `knowledge-base-for-startups`. - Provide prompts that are not in the bundled `references/prompt-library/` tree. Do not improvise one; point the user to the full library at , which carries prompts this plugin does not bundle. - Install the downloadable agents on the user's behalf. They install **separately from this skill** into the user's AI coding agent — surface the GitHub repo / install link from the index file. - Override the source-page disclaimer. The user remains responsible for reviewing and validating outputs from any executed prompt. ## Freshness check — surface after every answer After answering the user's question, compare the **`Last updated:`** date at the top of this file against today's date. If the gap is **more than 6 months**, append a short note to your reply suggesting the user refresh the skill: > _"This skill's content was last refreshed on ``, more than 6 months ago — some prompts and downloadable agents may have been added, removed, or revised. To pull the latest content, run:_ > > - _`npx skills update prompt-library-for-startups` — update the installed copy to the latest version_ > > _Then restart your AI agent so the new content is picked up."_ Do **not** show this note when the skill is fresh (≤6 months). Do **not** repeat it within a single conversation; once is enough.