ec2-spot-fleet-deployer
Provisions EC2 Spot Fleets with production defaults: launch template configuration, Spot Fleet request (target capacity, allocation strategy: lowestPrice/capacityOptimized/diversified/ priceCapacityOptimized), instance pools, fleet type (request/maintain), replacement strategies, capacity rebalance, Spot placement score, capacity reservations. Emits a READY_TO_DEPLOY checklist with verification commands. Use when creating a Spot Fleet request, choosing an allocation strategy, configuring instance diversification, enabling capacity rebalance, or integrating Spot Fleets with capacity reservations. Triggers: create spot fleet, EC2 Spot Fleet request, spot instance pool, allocation strategy, capacity rebalance, spot placement score, priceCapacityOptimized, lowestPrice spot, capacityOptimized spot, diversified spot fleet, maintain spot fleet.
- License 许可
- Apache-2.0
- 最近更新
- 2026-09-24
- Stars
- ★ 4
- Forks
- 0
01 前言 Front Matter
- name
- ec2-spot-fleet-deployer
- license
- Apache-2.0
- metadata
- { "tags": "aws, ec2, spot-fleet, cloudops, deploy, compute, provisioning, allocation-strategy, capacity-rebalance, launch-template, spot-instances", "phase": "1", "author": "Jacky Chan — AWS Community Builder", "domain": "aws-cloudops", "family": "Compute", "version": "0.1.0", "keywords": "aws, ec2, spot fleet, spot instances, cloudops, deploy, provisioning, allocation strategy, lowestPrice, capacityOptimized, diversified, priceCapacityOptimized, instance pools, target capacity, capacity rebalance, spot placement score, capacity reservations, launch template, fleet type, maintain", "task_type": "deploy", "complexity": "high", "entry_point": "false", "skill_class": "capability", "when_to_use": "Invoke when the user wants to create an EC2 Spot Fleet request (maintain or one-time request), select an allocation strategy (lowestPrice, capacityOptimized, diversified, priceCapacityOptimized), configure instance pool diversification, enable capacity rebalance and replacement, use Spot placement score for capacity awareness, or integrate Spot Fleets with capacity reservations. Do NOT invoke for single Spot Instance requests (use RunInstances with InstanceMarketOptions), for On-Demand Capacity Reservations standalone (use capacity-reservation operators), for EC2 Auto Scaling Groups with mixed instances (use autoscaling-policy-deployer), or for auditing existing Spot Fleet configurations (use autoscaling-group-auditor).", "dependencies": "aws-orchestrator", "requires_llm": "true", "verdict_shape": "READY_TO_DEPLOY | PREREQUISITES_MISSING", "lifecycle_status": "active", "supports_pipeline": "true" }
- description
- Provisions EC2 Spot Fleets with production defaults: launch template configuration, Spot Fleet request (target capacity, allocation strategy: lowestPrice/capacityOptimized/diversified/ priceCapacityOptimized), instance pools, fleet type (request/maintain), replacement strategies, capacity rebalance, Spot placement score, capacity reservations. Emits a READY_TO_DEPLOY checklist with verification commands. Use when creating a Spot Fleet request, choosing an allocation strategy, configuring instance diversification, enabling capacity rebalance, or integrating Spot Fleets with capacity reservations. Triggers: create spot fleet, EC2 Spot Fleet request, spot instance pool, allocation strategy, capacity rebalance, spot placement score, priceCapacityOptimized, lowestPrice spot, capacityOptimized spot, diversified spot fleet, maintain spot fleet.
- compatibility
- Agent runtime that reads SKILL.md (Claude Code, Cursor, Windsurf, Codex, Gemini). For live deployment: AWS CLI v2 with ec2, iam, and service-roles access. Works with Terraform aws_spot_fleet_request resources and CloudFormation AWS::EC2::SpotFleet templates.
02 资源文件 20
- 参考文档eval/test-cases.yaml5.8 KB
- 参考文档evals/baselines/capacity-rebalance-enabled.md795 B
- 参考文档evals/baselines/maintain-price-capacity-optimized.md890 B
- 参考文档evals/baselines/missing-iam-role.md763 B
- 参考文档evals/baselines/request-lowest-price-burst.md817 B
- 参考文档evals/baselines/spot-placement-score.md843 B
- 参考文档evals/eval_config.yaml1.7 KB
- 参考文档evals/evals.json11.1 KB
- 参考文档evals/prompts/capacity-rebalance-enabled.md655 B
- 参考文档evals/prompts/maintain-price-capacity-optimized.md689 B
- 参考文档evals/prompts/missing-iam-role.md444 B
- 参考文档evals/prompts/request-lowest-price-burst.md593 B
- 参考文档evals/prompts/spot-placement-score.md657 B
- 参考文档examples/README.md6.0 KB
- 参考文档references/advanced-patterns.md7.6 KB
- 参考文档references/allocation-strategies.md6.8 KB
- 参考文档references/diagnostic-commands.md791 B
- 参考文档references/error-handling.md1.1 KB
- 参考文档references/provisioning-cli-commands.md8.6 KB
- 参考文档skill.usf.yaml4.2 KB
03 SKILL.md 正文
EC2 Spot Fleet Deployer
An AWS CloudOps agent skill that provisions EC2 Spot Fleets with correct defaults. The skill walks the operator through fleet type selection, launch template configuration, allocation strategy choice, instance pool diversification, target capacity, replacement strategies, capacity rebalance, and Spot placement score. It captures capacity and cost decisions, explains why each default matters, and emits a READY_TO_DEPLOY checklist with copy-pasteable verification commands.
Activation keywords
create Spot Fleet, EC2 Spot Fleet request, spot allocation strategy, capacity rebalance spot, Spot placement score, priceCapacityOptimized, lowestPrice spot, capacityOptimized spot, diversified Spot Fleet, maintain Spot Fleet, Spot instance pool, target capacity, Spot Fleet launch template.
STRICT output contract
When this skill is invoked with a Spot-Fleet-provisioning request
(fleet type, allocation strategy, instance types, target capacity,
or a partial configuration), the agent MUST respond with the
READY_TO_DEPLOY checklist defined in the "Output format" section
using the literal all-caps labels SPOT_FLEET:, VERDICT:,
CHECKLIST:, and VERIFICATION_COMMANDS:. Do NOT preface the
checklist with prose, headings, or disclaimers — emit the block as
the first lines of the response. This contract is what assertion-
based evals and downstream provisioning pipelines rely on;
deviating from the literal labels breaks automation silently.
If any prerequisite is missing, the verdict is
PREREQUISITES_MISSING with a specific gap citation in the
checklist (marked [✗]), and READY_TO_DEPLOY MUST NOT also
appear.
Quick navigation
| Section | When to read |
|---|---|
| Prerequisites | Always — verify before provisioning |
| Step 1 — Fleet type (maintain vs request) | Fleet lifecycle decision |
| Step 2 — Launch template configuration | AMI, instance type, network |
| Step 3 — Allocation strategy | Cost vs capacity trade-off |
| Step 4 — Instance pools and diversification | Interruption resilience |
| Step 5 — Target capacity (units vs vCPUs) | Sizing |
| Step 6 — Replacement strategies | Interruption handling |
| Step 7 — Capacity rebalance | Proactive rebalancing |
| Step 8 — Spot placement score | Capacity-aware launch |
| Step 9 — Capacity reservations integration | Guaranteed capacity |
| Step 10 — Recent features | Latest |
| NEVER do these things | Review before signing off |
| Output format | The literal checklist template |
| references/allocation-strategies.md | Strategy deep dive |
| references/provisioning-cli-commands.md | Copy-pasteable CLI sequence |
Mindset
Full mindset and the three provisioning misconceptions moved to references/advanced-patterns.md.
Configuration dependency graph (novel heuristic)
Full dependency table and cross-dependency gotchas moved to references/advanced-patterns.md.
Expert heuristic: allocation strategy selection
Workload-profile decision framework and key implications moved to references/allocation-strategies.md.
Expert heuristic: capacity rebalance and replacement
Rebalance-vs-default-replacement comparison, lifecycle, and enabling JSON moved to references/advanced-patterns.md.
Prerequisites (verify before provisioning)
Before emitting provisioning commands, verify these prerequisites. If any are missing, the verdict is PREREQUISITES_MISSING with a specific gap citation.
| Prerequisite | Why it matters | How to verify |
|---|---|---|
| AWS account with EC2 Spot Fleet access | Cannot provision without it | aws sts get-caller-identity |
| Region selected | Instance types and AZ availability vary by region | aws configure get region |
| IAM service role for Spot Fleet | Spot Fleet needs AWSServiceRoleForEC2SpotFleet to launch instances |
aws iam get-role --role-name AWSServiceRoleForEC2SpotFleet |
| Launch template exists (or create one) | Fleet needs a standardized instance configuration | aws ec2 describe-launch-templates |
| AMI valid in the region | AMIs are region-scoped; cross-region AMI IDs are invalid | aws ec2 describe-images --image-ids <ami-id> |
| Instance types available in the region | Not all instance types are available in all regions/AZs | aws ec2 describe-instance-types --instance-types <type> |
| At least 2-3 instance types specified | Single-type fleets have no diversification fallback | List in the Overrides |
| At least 2 AZs specified | Single-AZ fleets lose all capacity on AZ failure | List AZs in Overrides |
| Target capacity identified | Determines fleet size; maintain fleets auto-adjust |
Positive integer |
| Spot placement score (optional) | Pre-check capacity availability before creating fleet | aws ec2 get-spot-placement-scores |
If any prerequisite is missing, output VERDICT: PREREQUISITES_MISSING
and cite the specific gap.
Step 1 — Fleet type (maintain vs request)
The first decision is whether the fleet should continuously maintain its target capacity or fill it once and stop.
Decision tree:
Does the workload need continuous capacity (long-running, stateless)?
├── YES → Fleet type: maintain
│ AWS continuously monitors and replaces interrupted instances.
│ Target capacity can be modified (scale up/down).
│ Capacity rebalance can be enabled.
│ Use for: web servers, batch processing, CI/CD workers
└── NO → Is this a one-time burst (job queue, HPC, data processing)?
├── YES → Fleet type: request
│ Fleet fills once and stops. No replacement.
│ Target capacity CANNOT be modified after creation.
│ Capacity rebalance is silently ignored.
│ Use for: one-time HPC jobs, data migration, burst compute
└── NO → Fleet type: maintain (the safe default)
Feature comparison:
| Feature | maintain | request |
|---|---|---|
| Replacement on interruption | Yes (automatic) | No |
| Target capacity modification | Yes (live) | No (immutable) |
| Capacity rebalance | Supported | Not supported (ignored) |
| Excess capacity termination | Configurable | N/A |
| Expiry | Until cancelled | Fills once, then done |
| Best for | Long-running workloads | One-time burst jobs |
Common mistake: using request for a web tier and expecting
AWS to replace interrupted instances. It will not — the fleet fills
once and stops. Use maintain.
Step 2 — Launch template configuration
A launch template defines the instance configuration: AMI, instance type, key pair, security groups, IAM instance profile, user data, and storage. The Spot Fleet references the launch template and overrides instance type and AZ via the Overrides array.
Create a launch template:
aws ec2 create-launch-template \
--launch-template-name spot-fleet-web-server \
--launch-template-data '{
"ImageId": "ami-0abcdef1234567890",
"InstanceType": "m5.large",
"KeyName": "my-key-pair",
"SecurityGroupIds": ["sg-0abc123"],
"IamInstanceProfile": { "Arn": "arn:aws:iam::123456789012:instance-profile/spot-fleet-role" },
"UserData": "'$(base64 user-data.sh)'",
"BlockDeviceMappings": [{
"DeviceName": "/dev/xvda",
"Ebs": { "VolumeSize": 30, "VolumeType": "gp3", "DeleteOnTermination": true }
}],
"TagSpecifications": [{
"ResourceType": "instance",
"Tags": [
{ "Key": "Name", "Value": "spot-fleet-web" },
{ "Key": "Environment", "Value": "production" }
]
}]
}'
Key rules:
- The instance type in the launch template is the DEFAULT; the Spot Fleet Overrides override it per pool.
- The AMI must be valid in the fleet's region.
- EBS volumes should be
DeleteOnTermination: truefor Spot (otherwise orphaned volumes accumulate cost). - Tag the instances at launch (launch-template tags are more reliable than Spot Fleet-level tags).
Step 3 — Allocation strategy
See the "Expert heuristic: allocation strategy selection" section above for the decision framework. Summary:
| Strategy | Picks pools based on | Diversification | Interruption risk | Cost |
|---|---|---|---|---|
priceCapacityOptimized |
Price + capacity availability (RECOMMENDED) | Implicit (multi-pool) | Low-Medium | Low |
capacityOptimized |
Capacity availability (max free capacity) | Implicit (multi-pool) | Lowest | Medium |
lowestPrice |
Lowest price per AZ | None (concentrates) | Highest | Lowest |
diversized |
Even spread across all pools | Explicit (equal) | Low | Medium |
Setting the allocation strategy:
{
"AllocationStrategy": "priceCapacityOptimized",
"TargetCapacitySpecification": {
"TotalTargetCapacity": 10,
"DefaultTargetCapacityType": "spot"
}
}
Common mistake: using lowestPrice with a single instance type.
The fleet concentrates in one pool and is interrupted all at once.
Always use priceCapacityOptimized or capacityOptimized unless you
have a specific cost-optimization requirement with many (10+) instance
types.
Step 4 — Instance pools and diversification
Instance pools are unique combinations of instance type + AZ. The Spot Fleet draws from these pools based on the allocation strategy. More pools = better diversification = lower interruption risk.
Overrides specify the pools (9 pools = 3 types x 3 AZs):
"LaunchTemplateConfigs": [{
"LaunchTemplateSpecification": { "LaunchTemplateId": "lt-0abc123", "Version": "1" },
"Overrides": [
{ "InstanceType": "m5.large", "AvailabilityZone": "us-east-1a" },
{ "InstanceType": "m5.large", "AvailabilityZone": "us-east-1b" },
{ "InstanceType": "m5.large", "AvailabilityZone": "us-east-1c" },
{ "InstanceType": "m5a.large", "AvailabilityZone": "us-east-1a" },
{ "InstanceType": "c5.large", "AvailabilityZone": "us-east-1b" },
{ "InstanceType": "c5.large", "AvailabilityZone": "us-east-1c" }
]
}]
Diversification best practices:
- Specify at least 2-3 instance types (e.g., m5, m5a, c5).
- Specify at least 3 AZs (all AZs in the region).
- This gives 9+ pools, dramatically reducing interruption risk.
- Use instance type families with similar vCPU/memory ratios so your workload scales predictably.
- Consider using
InstanceRequirements(attribute-based) instead of specific types for flexible selection.
Attribute-based instance selection (InstanceRequirements):
"Overrides": [
{
"InstanceRequirements": {
"VCpuCount": { "Min": 2, "Max": 4 },
"MemoryMiB": { "Min": 8192 },
"AcceleratorCount": { "Max": 0 },
"BurstablePerformance": "excluded"
},
"AvailabilityZone": "us-east-1a"
}
]
This lets AWS pick any instance type matching the requirements, maximizing pool count.
Step 5 — Target capacity (units vs vCPUs vs instances)
Target capacity can be measured in three ways:
| Unit type | How it works | When to use |
|---|---|---|
units |
Weighted: each override has a WeightedCapacity value |
Mixed-size fleets (some instances count as 2 units, others as 1) |
vcpu |
Each instance counts as its vCPU count | vCPU-based capacity planning |
instances (default) |
Each instance counts as 1 | Simple equal-weight fleets |
"TargetCapacitySpecification": {
"TotalTargetCapacity": 20,
"DefaultTargetCapacityType": "spot",
"TargetCapacityUnitType": "vcpu"
}
Common mistake: using instances with mixed instance types
where some are 2x the size of others. The fleet thinks it needs 20
instances, but if it picks large instances, you have way more
capacity than intended. Use vcpu or units for mixed fleets.
Step 6 — Replacement strategies (InstanceInterruptionBehavior)
When a Spot Instance is interrupted, the fleet's behavior depends
on InstanceInterruptionBehavior:
| Behavior | What happens to the interrupted instance | When to use |
|---|---|---|
terminate (default) |
Instance is terminated (EBS volumes deleted if DeleteOnTermination) | Stateless workloads |
stop |
Instance is stopped (can be restarted later) | Stateful workloads that can resume |
hibernate |
Instance is hibernated (RAM saved to disk) | Long-running jobs with checkpointing |
{
"InstanceInterruptionBehavior": "terminate",
"SpotMaintenanceStrategies": {
"CapacityRebalance": {
"ReplacementStrategy": "launch"
}
}
}
Key rule: InstanceInterruptionBehavior is set at the Spot
Fleet request level and applies to ALL instances in the fleet. It
cannot be overridden per-instance after launch.
Step 7 — Capacity rebalance
See the "Expert heuristic: capacity rebalance and replacement" section above for the full lifecycle. Summary configuration:
"SpotMaintenanceStrategies": {
"CapacityRebalance": {
"ReplacementStrategy": "launch"
}
}
Replacement strategies:
launch(default): Fleet launches a new replacement instance BEFORE terminating the old one. This is the recommended strategy.terminate: Fleet terminates the old instance immediately and then launches a replacement. Higher risk of capacity gap.
Requirements:
- Fleet type MUST be
maintain. - Fleet MUST have
InstanceInterruptionBehaviorset toterminateorstop. - The replacement instance is launched from the same launch template + overrides.
Common mistake: enabling capacity rebalance on a request fleet.
It is silently ignored. Only maintain fleets support rebalancing.
Step 8 — Spot placement score
Spot placement score is a pre-flight capacity check. Before creating a fleet, query the score to see which instance type + AZ combinations have the best capacity availability.
Query command and sample output moved to references/diagnostic-commands.md.
Score interpretation:
- Score 1-3: Low availability. High interruption risk. Avoid.
- Score 4-7: Medium availability. Acceptable for fault-tolerant.
- Score 8-10: High availability. Low interruption risk. Prefer.
Key implication: Spot placement score is ADVISORY ONLY. It does not reserve capacity. The fleet may still be interrupted in a high-scoring pool if overall Spot demand spikes.
When to use: Before creating large fleets (>100 instances), when choosing between instance types, or before scaling up.
Step 9 — Capacity reservations integration
On-Demand Capacity Reservations (ODCR) can be used alongside Spot Fleets. When the fleet launches in an AZ+instance-type combination that has an open ODCR, the ODCR is consumed first (guaranteed capacity, On-Demand billing), then Spot fills the rest.
Configuration for ODCR preference:
{
"LaunchTemplateConfigs": [{
"LaunchTemplateSpecification": { "LaunchTemplateId": "lt-0abc123", "Version": "1" },
"Overrides": [{
"InstanceType": "m5.large",
"AvailabilityZone": "us-east-1a"
}]
}],
"TargetCapacitySpecification": {
"TotalTargetCapacity": 10,
"OnDemandTargetCapacity": 3,
"DefaultTargetCapacityType": "spot"
},
"OnDemandOptions": {
"CapacityReservationOptions": {
"UsageStrategy": "use-capacity-reservations-first"
}
}
}
Key rules:
- ODCR must match instance type + AZ + platform exactly.
OnDemandTargetCapacityspecifies how many units to run as On-Demand (consumed from ODCR if available).- The rest of the fleet runs as Spot.
- This gives a hybrid: guaranteed capacity for critical instances + cost savings for the rest.
Step 10 — Recent features
Recent AWS feature notes (2023-2026) moved to references/advanced-patterns.md.
NEVER do these things
-
NEVER use
lowestPricewith a single instance type. The fleet concentrates in one pool. When that pool is reclaimed, the entire fleet is interrupted simultaneously with no fallback. UsepriceCapacityOptimizedwith at least 3 instance types. -
NEVER use
requestfleet type for long-running workloads. Arequestfleet fills once and stops. It does NOT replace interrupted instances. Usemaintainfor any workload that needs continuous capacity. -
NEVER specify only one instance type or one AZ. Single-type or single-AZ fleets have no diversification. Always specify 2-3+ instance types across 3+ AZs for interruption resilience.
-
NEVER enable capacity rebalance on a
requestfleet. It is silently ignored. Capacity rebalance only works withmaintainfleet type. Verify fleet type before configuring rebalance. -
NEVER treat Spot placement score as a capacity guarantee. It is advisory only. A score of 10 does not reserve capacity. Use it to avoid low-scoring pools, not as a reservation.
Output format
SPOT_FLEET: <fleet-id> (allocation strategy: <strategy>)
VERDICT: READY_TO_DEPLOY | PREREQUISITES_MISSING
CHECKLIST:
[✓|✗] Fleet type: maintain | request
[✓|✗] Allocation strategy: priceCapacityOptimized | capacityOptimized | lowestPrice | diversified
[✓|✗] Launch template: <lt-id> (version <n>)
[✓|✗] Instance types: <type-1>, <type-2>, <type-3> (in <n> AZs = <pool-count> pools)
[✓|✗] Target capacity: <n> <unit-type> (spot: <n>, on-demand: <n>)
[✓|✗] InstanceInterruptionBehavior: terminate | stop | hibernate
[✓|✗] Capacity rebalance: enabled (launch strategy) | disabled
[✓|✗] Spot placement score: checked (top pool score: <score>) | not checked
[✓|✗] Capacity reservations (ODCR): used-first (<n> On-Demand units) | not used
[✓|✗] IAM service role: AWSServiceRoleForEC2SpotFleet
[✓|✗] Tags: <key=value list>
VERIFICATION_COMMANDS:
aws ec2 describe-spot-fleet-requests --spot-fleet-request-ids <fleet-id>
aws ec2 describe-spot-fleet-instances --spot-fleet-request-id <fleet-id>
aws ec2 describe-spot-fleet-request-history --spot-fleet-request-id <fleet-id>
Worked example — maintain fleet with priceCapacityOptimized and rebalance
SPOT_FLEET: sfr-1234567890abcdef0 (allocation strategy: priceCapacityOptimized)
VERDICT: READY_TO_DEPLOY
CHECKLIST:
[✓] Fleet type: maintain
[✓] Allocation strategy: priceCapacityOptimized
[✓] Launch template: lt-0abc123def456 (version 1)
[✓] Instance types: m5.large, m5a.large, c5.large (in 3 AZs = 9 pools)
[✓] Target capacity: 20 vcpu (spot: 20, on-demand: 0)
[✓] InstanceInterruptionBehavior: terminate
[✓] Capacity rebalance: enabled (launch strategy)
[✓] Spot placement score: checked (top pool score: 10)
[✓] Capacity reservations (ODCR): not used
[✓] IAM service role: AWSServiceRoleForEC2SpotFleet
[✓] Tags: Environment=production, Service=batch-processing
VERIFICATION_COMMANDS:
aws ec2 describe-spot-fleet-requests --spot-fleet-request-ids sfr-1234567890abcdef0
aws ec2 describe-spot-fleet-instances --spot-fleet-request-id sfr-1234567890abcdef0
aws ec2 describe-spot-fleet-request-history --spot-fleet-request-id sfr-1234567890abcdef0
Error handling
Fleet failure modes and their fixes moved to references/error-handling.md.
References (load on demand)
- references/advanced-patterns.md — mindset misconceptions, configuration dependency graph, capacity rebalance deep dive, recent AWS features (moved from this file)
- references/allocation-strategies.md — allocation-strategy workload-profile heuristic (moved from this file) plus the strategy comparison matrix and decision framework
- references/provisioning-cli-commands.md — full copy-pasteable provisioning CLI sequence
- references/diagnostic-commands.md — Spot placement score query and sample output (moved from this file)
- references/error-handling.md — fleet stuck/partially filled, rebalance not working, MaxSpotInstanceCountExceeded, immutable allocation strategy (moved from this file)
Domain
AWS CloudOps / EC2 Spot Fleet Provisioning & Cost-Optimized Compute Capacity Management.
AWS documentation
- Spot Fleet User Guide — https://docs.aws.amazon.com/AWSEC2/latest/UserGuide/spot-fleet.html
- Allocation strategies — https://docs.aws.amazon.com/AWSEC2/latest/UserGuide/spot-fleet-allocation-strategy.html
- Capacity rebalance — https://docs.aws.amazon.com/AWSEC2/latest/UserGuide/capacity-rebalance.html
- Spot placement score — https://docs.aws.amazon.com/AWSEC2/latest/UserGuide/spot-placement-score.html
- Spot Fleet request options — https://docs.aws.amazon.com/AWSEC2/latest/UserGuide/spot-request-options.html
- Instance interruption behavior — https://docs.aws.amazon.com/AWSEC2/latest/UserGuide/spot-interruptions.html
- On-Demand Capacity Reservations — https://docs.aws.amazon.com/AWSEC2/latest/UserGuide/ec2-capacity-reservations.html
- Attribute-based instance selection — https://docs.aws.amazon.com/AWSEC2/latest/UserGuide/spot-instance-request-types.html#attribute-based-instance-type-selection
04 同仓库更多 Skills
xray-tracing-deployer
Deploys AWS X-Ray distributed tracing with production-grade configuration: X-Ray daemon deployment (EC2 systemd, ECS sidecar, EKS DaemonSet, Lambda built-in), SDK instrumentation (Java, Python, Node.js, Go, .NET) with annotations and metadata, sampling rules (default reservoir + rate vs custom rules with URL / method / service-name predicates), service map generation, trace query and retrieval, groups and insights, CloudWatch ServiceLens integration, IAM permissions (PutTraceSegments, GetSamplingRules), and latest features (Lambda Powertools tracing, OpenTelemetry / ADOT Collector support, W3C trace context). Emits a READY_TO_DEPLOY checklist with every configuration item verified. Use when enabling X-Ray on a new application, adding tracing to ECS / EKS / Lambda, validating a sampling configuration, or generating daemon deployment CLI and IaC templates. Triggers: X-Ray, distributed tracing, X-Ray daemon, sampling rules, service map, X-Ray SDK, OpenTelemetry, ADOT, Lambda Powertools, CloudWatch ServiceLens.
wellarchitected-workload-auditor
Audits AWS Well-Architected Tool workloads for review staleness, per-pillar high-risk issue counts, milestone tracking gaps, and remediation plan completeness. Emits a deterministic verdict (STALE_REVIEW | HIGH_RISK | CONFIG_GAP | OK) per workload with per-pillar risk breakdown, milestone trend analysis, and specific remediation. Use when reviewing Well-Architected workloads, checking WA review freshness, auditing high-risk issues per pillar, validating milestone coverage, or tracking remediation plans.
wellarchitected-review-operator
Operates AWS Well-Architected Tool reviews end-to-end — creates workloads, runs pillar reviews (operational excellence, security, reliability, performance, cost optimization, sustainability, Prosperity), answers questions with risk-tier rationale, generates reports and improvement plans, creates milestones, and integrates with Trusted Advisor and Well-Architected Labs. Runs deterministic pre-checks (workload existence, lens availability, pillar coverage), emits the exact update-answer / get-consolidated-report CLI behind a CONFIRM gate, and verifies risk-tier transitions. Emits READY | BLOCKED | COMPLETED. Use when creating a workload, answering pillar questions, generating an improvement plan, creating a milestone, or integrating with Trusted Advisor. Triggers: well-architected tool, well architected review, workload review, pillar, operational excellence, security, reliability, performance, cost optimization, sustainability, prosperity, improvement plan, milestone, trusted advisor.
wafv2-web-acl-deployer
Provisions AWS WAFv2 Web ACLs with correct production defaults: scope selection (CLOUDFRONT vs REGIONAL), managed rule groups (AWSManagedRulesCommonRuleSet, SQLiRuleSet, LinuxRuleSet, WindowsRuleSet, IPReputationList, BotControlRuleSet, AmazonIpReputationList, ATP), custom rules (byte match, IP set, regex pattern, geo match, size constraint, rate-based), rule priority ordering, logging destinations (Kinesis Firehose, CloudWatch Logs, S3), resource association (ALB, API Gateway, CloudFront), rate-based rules with aggregate keys (IP, forwarded IP), Challenge and CAPTCHA actions, and account takeover prevention. Emits a READY_TO_DEPLOY checklist. Use when creating a Web ACL, attaching managed rules, configuring rate limiting, enabling WAF logging, associating a Web ACL with an ALB/API Gateway/CloudFront, or configuring CAPTCHA/Challenge actions. Triggers: create WAF Web ACL, WAFv2, managed rule groups, WAF rate limiting, WAF logging, WAF CAPTCHA, WAF Challenge action, bot control, ATP managed rules.
wafv2-web-acl-auditor
Audits AWS WAFv2 Web ACL configurations to determine whether the ACL provides effective protection: default-action posture (Allow vs Block), managed-rule-group coverage gaps, rule effectiveness (BLOCK vs COUNT, shadow/bypass rules, stale exclusions), rate-based rule correctness, logging and visibility configuration, and text-transformation bypass vectors. Emits a deterministic verdict per Web ACL. Use when reviewing a WAFv2 Web ACL for security posture, checking managed-rule coverage, validating rate-limit configuration, auditing WAF logging, or hardening a WAF before production deployment. Triggers: WAF, WAFv2, Web ACL, managed rule group, AWSManagedRules, rate-based rule, CAPTCHA, Challenge, OverrideAction, Count mode, forwarded IP, Bot Control, ATP, text transformation, WAF logging, WAF visibility, OWASP, SQLi, XSS, RFI.
waf-rule-deployer
Provisions AWS WAFv2 rule sets with production defaults: Web ACL creation (CloudFront scope in us-east-1 vs Regional scope), rule groups (managed vs custom) with capacity planning in WCU (Web ACL Capacity Units, 1500 default per ACL), managed rule groups (AWSManagedRulesCommonRuleSet, AWSManagedRulesSQLiRuleSet, AWSManagedRulesAmazonIpReputationList, AWSManagedRulesLinuxRuleSet, AWSManagedRulesWindowsRuleSet, AWSManagedRulesUnixRuleSet), custom rules (byte-match, regex pattern set, geo-match, rate-based), IP set and regex pattern set resources, rule priority ordering (lowest number evaluated first), action types (allow, block, count, CAPTCHA, Challenge), bot control managed rule group, account takeover. Triggers: create WAF rule, WAF managed rule group, WAF custom rule, WAF byte match, WAF geo match, WAF rate based rule, WAF IP set, WAF regex pattern set, WAF WCU budget, WAF CloudFront ACL, WAF logging Kinesis Firehose, WAF bot control, WAF ATP, WAF label matching, WAF Challenge action, WAF CAPTCHA.