AWS Media Services

Use this skill when the user is doing ANY of the following with AWS media services:

• Building or modifying a video workflow (ingest, encoding, packaging, ad insertion, CDN delivery)
• Reviewing, evaluating, or auditing an existing or proposed configuration for best practices, compliance, or correctness
• Troubleshooting errors, alerts, or unexpected behavior
• Asking questions about service constraints, compatibility rules, encoder settings, IAM policies, security groups, DRM, SCTE-35, caching, scaling, or any other AWS media topic
• Planning a new workflow and needs to understand requirements, dependencies, or the discovery checklist
• Optimizing cost, resilience, or performance of an existing workflow

If the user mentions ANY AWS media service by name (MediaConnect, MediaLive, MediaConvert, MediaPackage, MediaTailor) or related concepts (HLS, DASH, CMAF, ABR ladder, SCTE-35, epoch locking, origin shield, CDI, SRT, RTMP, transcode, live streaming, VOD pipeline, ad insertion, SSAI), load this skill.

This file is the orchestrator. It tells you which reference files to load based on the user’s task. Do not load all files at once. Load on demand per the routing table below.

MANDATORY FIRST STEP — Read Before Doing Anything Else

STOP. Before you make ANY API call, pull ANY resource configuration, write ANY code, or produce ANY recommendation, you MUST load the required rule files into context. No exceptions. This applies to every task type: building, modifying, evaluating, troubleshooting, or answering questions.

The rule content is split across:

• references/service-rules-core.md — cross-service compatibility constraints, common agent mistakes, and security considerations. ALWAYS loaded.
• references/service-rules-encoding.md — shared encoder settings (QVBR, GOP, profiles, quality tuning, deinterlacing). Loaded for any encoding task.
• references/service-rules-mediaconnect.md
• references/service-rules-medialive.md
• references/service-rules-mediaconvert.md
• references/service-rules-mediapackage.md
• references/service-rules-mediatailor.md
• references/service-rules-cloudfront.md

Per-service rule files are loaded conditionally based on which services the query involves. Without the relevant rules loaded, you will produce recommendations based on general knowledge that may contradict the documented rules, flag correct configurations as problems, or miss critical constraints entirely.

Sequence (every time):

1. Read references/service-rules-core.md — confirm it is in context before proceeding. ALWAYS required.
2. Determine the task type and consult the Task Routing tables below to identify both the workflow/troubleshooting file AND the per-service rules to co-load.
3. For ad-hoc questions that don’t match a workflow or troubleshooting category: if the user names specific service(s), load those service rule files. If the question is symptom-based or you cannot determine which services are involved, load ALL per-service rule files.
4. Only then begin resource discovery, API calls, or implementation.

If you find yourself about to call aws medialive describe-channel or similar and you have NOT yet read the core rules file plus the relevant per-service rules, stop and go back to step 1.

Task Routing

Building or modifying a workflow

Determine the workflow type. Load the matching workflow file AND the per-service rules listed in the right column (the core rules file is already loaded from the mandatory first step above).

Troubleshooting

Load the relevant troubleshooting file(s) for the problem area AND the per-service rules listed in the right column. The core rules file is already loaded from the mandatory first step above. Load only the files needed.

Quick reference

For a quick answer about a specific service’s constraints, valid parameter values, or compatibility rules:

1. references/service-rules-core.md is already loaded (mandatory first step).
2. If the user names specific service(s) in the query, load those per-service rule files (e.g. references/service-rules-medialive.md).
3. If the question crosses service boundaries (e.g., MediaPackage→MediaTailor compatibility), the cross-service rules are already in service-rules-core.md.
4. If you can’t determine which services are involved, load ALL per-service rule files.

This includes the following topics:

• MediaTailor scaling patterns (multi-region, CDN origin shield, prefetch)
• MediaConnect Router (paired resilient zonal endpoints) and downstream NDI converter caveats
• Epoch locking and cross-region failover
• Contribution transport (SRT tuning, Direct Connect sizing, RTP+FEC)
• Multi-CDN architecture patterns
• CloudFront advanced patterns (Embedded POPs / ePOP, CF Functions, segment sharding)
• Operational readiness for live events (capacity planning, KPI baselines)
• DRM production patterns and MediaPackage v1→v2 migration
• Audio, HDR, and codec configuration details

Workflow Discovery Checklist

Before creating ANY resources for a new workflow, gather requirements from the user in a fixed order (mode → cost/scale → approach → source → network → encoding → ad insertion → packaging → IAM → VOD-specific). Do not proceed until all required items are answered; later questions depend on earlier answers.

Load references/workflow-discovery-checklist.md for the full ordered question set whenever building or modifying a workflow.

Plan Validation (before presenting to user)

Before presenting a proposed configuration to the user, cross-check against the loaded service rules:

• Codec/container compatibility (e.g., HEVC requires fMP4, not TS)
• GOP/segment duration alignment across all ABR rungs
• Dependency ordering (downstream before upstream)
• IAM trust policy includes all required service principals + sts:TagSession
• IAM permissions scoped to specific resource ARNs (no Resource: “*”)
• Security posture (CIDR scoped, OAC configured, no Principal: *)
• All prerequisite resources included (IAM roles, VPC, MediaPackage channel MUST be in the plan before resources that depend on them)

Do NOT make API calls during this step. This is a rules-based review of the proposed plan against constraints already in context.

Agents MUST NOT skip dependency steps to save time. All prerequisites (IAM, VPC, MediaPackage channel, etc.) MUST be created before resources that reference them. Skipping steps and fixing errors reactively leads to circular troubleshooting loops. Execute the plan in dependency order.

Resource Creation Order

Respect dependency order — a resource must exist before anything that references it (live chain, VOD chain, live-to-VOD addition, and the common flow patterns table).

Load references/resource-creation-order.md for the full dependency chains and flow-pattern table when creating resources.

Startup and Shutdown (Live Only)

Start upstream first and poll each resource to its ready state before proceeding (STARTING ≠ running); shut down in reverse (STOPPING ≠ stopped). Poll on a tight ~5s cadence.

Load references/startup-shutdown.md for the ordered start/stop tables and the polling pattern (including MediaConnect Router lifecycle) when operating a live workflow.

Tooling

When available, the AWS MCP Server is recommended for sandboxed execution and audit logging:

• AWS MCP Server (aws-mcp-server): AWS API access, Agent SOPs for guided multi-step workflows, and documentation search via search_documentation and read_documentation.
• AWS IaC MCP Server (aws-iac-mcp-server): CloudFormation template validation, CDK documentation search, CDK code samples, and IaC best practices.

Large response warning: MediaLive describe-channel and MediaPackage list-* calls return very large JSON. Always use --query (JMESPath) to extract only the fields you need.

# Good — extract only what you need
aws medialive describe-channel --channel-id 12345 --query "{State:State,Name:Name,Id:Id}"
aws medialive list-channels --query "Channels[].{Name:Name,Id:Id,State:State}"
aws mediapackagev2 list-origin-endpoints --channel-group-name X --channel-name Y \
  --query "Items[].{Name:OriginEndpointName,Url:HlsManifests[0].Url}"