securityhub-control-compliance-auditor
Audits AWS Security Hub control compliance findings and maps each to a deterministic compliance verdict (FAILED | WARNING | PASSED | NOT_APPLICABLE), then maps every FAILED / WARNING control to a specific fix action. Handles the full Security Hub finding lifecycle — active findings, suppressed findings, resolved-but-still-failing, archived findings, NOT_AVAILABLE with StatusReasons, multi-account aggregation, and cross-standard control families (FSBP, CIS, PCI-DSS, NIST 800-53). Use when reviewing Security Hub control status, triaging a compliance gap report, mapping findings to remediation runbooks, or preparing for an audit / executive compliance summary.
- License 许可
- Apache-2.0
- 最近更新
- 2026-09-24
- Stars
- ★ 4
- Forks
- 0
01 前言 Front Matter
- name
- securityhub-control-compliance-auditor
- license
- Apache-2.0
- metadata
- { "tags": "aws, securityhub, security, compliance, fsbp, cis, pci-dss, nist, audit, remediation, asff", "phase": "2", "author": "Jacky Chan — AWS Community Builder", "domain": "aws-cloudops", "family": "Security", "version": "0.2.0", "keywords": "Security Hub, compliance, control status, FSBP, Foundational Security Best Practices, CIS Benchmark, PCI-DSS, NIST 800-53, ASFF, finding lifecycle, suppression, remediation runbook, compliance verdict, NOT_AVAILABLE, StatusReasons, Workflow Status, Record State, delegated administrator", "complexity": "high", "entry_point": "false", "when_to_use": "Reviewing Security Hub control compliance status, triaging a compliance gap report, mapping findings to remediation actions, auditing finding lifecycle states (suppressed / resolved / archived), preparing for a compliance audit, or generating an executive compliance summary across FSBP / CIS / PCI-DSS / NIST standards.", "requires_llm": "true", "verdict_shape": "FAILED | WARNING | PASSED | NOT_APPLICABLE", "supports_pipeline": "true" }
- description
- Audits AWS Security Hub control compliance findings and maps each to a deterministic compliance verdict (FAILED | WARNING | PASSED | NOT_APPLICABLE), then maps every FAILED / WARNING control to a specific fix action. Handles the full Security Hub finding lifecycle — active findings, suppressed findings, resolved-but-still-failing, archived findings, NOT_AVAILABLE with StatusReasons, multi-account aggregation, and cross-standard control families (FSBP, CIS, PCI-DSS, NIST 800-53). Use when reviewing Security Hub control status, triaging a compliance gap report, mapping findings to remediation runbooks, or preparing for an audit / executive compliance summary.
- compatibility
- Agent runtime that reads SKILL.md (Claude Code, Cursor, Windsurf, Codex, Gemini). No AWS CLI required for offline classification — the skill reasons over provided Security Hub finding JSON (ASFF). For live-account audits, AWS CLI v2 with securityhub access (ListEnabledStandards, GetEnabledStandards, GetFindings, batch-update-findings) and SSO or key-based credentials.
02 资源文件 21
- 参考文档eval/test-cases.yaml7.2 KB
- 参考文档evals/baselines/archived-finding.md715 B
- 参考文档evals/baselines/iam-password-policy-passed.md612 B
- 参考文档evals/baselines/not-available-no-resources.md651 B
- 参考文档evals/baselines/not-available-service-disabled.md655 B
- 参考文档evals/baselines/s3-bpa-failed.md759 B
- 参考文档evals/baselines/suppressed-failed.md706 B
- 参考文档evals/eval_config.yaml2.0 KB
- 参考文档evals/evals.json4.8 KB
- 参考文档evals/prompts/archived-finding.md889 B
- 参考文档evals/prompts/iam-password-policy-passed.md799 B
- 参考文档evals/prompts/not-available-no-resources.md850 B
- 参考文档evals/prompts/not-available-service-disabled.md850 B
- 参考文档evals/prompts/s3-bpa-failed.md841 B
- 参考文档evals/prompts/suppressed-failed.md896 B
- 参考文档example/end-to-end-audit.md7.8 KB
- 参考文档references/advanced-patterns.md18.5 KB
- 参考文档references/diagnostic-commands.md3.0 KB
- 参考文档references/error-handling.md2.4 KB
- 参考文档references/remediation-guidance.md8.0 KB
- 参考文档skill.usf.yaml4.1 KB
03 SKILL.md 正文
Security Hub Control-Compliance Auditor
Structural pattern: ASFF Audit Pipeline — Activation → Output Contract → Mindset → Decision Tree → Process Steps → Reference Tables → Anti-Patterns → Safety Checks. Read top-to-bottom for a cold start; jump to the Decision Tree for a single finding.
Quick Start (the 3 rules that prevent 80% of misclassifications)
- NOT_AVAILABLE is NOT the same as NOT_APPLICABLE. Check
StatusReasons[].Code:NO_RESOURCES/DISABLED_CONTROL→ NOT_APPLICABLE; everything else (especiallySUPPORTED_SERVICE_NOT_ENABLED) → WARNING. - SUPPRESSED + FAILED is WARNING, never PASSED. Suppression hides the finding but the control is still failing.
- ARCHIVED is always NOT_APPLICABLE. The resource was deleted or the control superseded — historical state, not current posture.
Full classification: see Decision Tree below. Detailed reasoning: see Process steps. Fix actions: see Control-to-fix-action mapping.
Activation
Moved to references/advanced-patterns.md. Four ASFF detection signals and the do-NOT-invoke exclusions (CloudTrail events, bare IAM policy JSON, Audit Manager, Config compliance JSON).
Output Contract (CRITICAL — read before any classification)
Every finding MUST produce this EXACT output block. An agent that omits any field fails the audit. Copy this template verbatim and fill the placeholders:
CONTROL: <ControlId from GeneratorId, e.g. S3.2>
VERDICT: <FAILED | WARNING | PASSED | NOT_APPLICABLE>
REASON: <Rule number + the specific field values that drove the verdict.
MUST cite Compliance.Status, Workflow.Status, RecordState, and StatusReasons
code if applicable. For SUPPRESSED findings, MUST include whether a Workflow
Note is present and its content or absence. 1-2 sentences.>
SEVERITY: <CRITICAL | HIGH | MEDIUM | LOW | INFORMATIONAL>
REMEDIATION: <Specific fix action from the control-to-fix-action table, or
"None required — control is compliant" for PASSED, or
"None required — control does not apply" for NOT_APPLICABLE.>
Rules for filling the template:
- ALWAYS include ALL FIVE headers (
CONTROL,VERDICT,REASON,SEVERITY,REMEDIATION) even when the finding is PASSED or NOT_APPLICABLE — a blankREMEDIATION: None requiredis correct, but omitting the header is a contract violation. - The
VERDICTvalue must be ONE of exactly four tokens:FAILED,WARNING,PASSED,NOT_APPLICABLE. Never output compound statuses likeWARNING (PENDING)orNOT_APPLICABLE/ARCHIVED. - For SUPPRESSED findings, the REASON field MUST mention the suppression
context: whether
Workflow.Noteis present, its author/date, or explicitly state "Suppression has no documented justification." Omitting the suppression context from a SUPPRESSED finding is a contract violation. - Never prefix the block with conversational text ("Sure, here is the
audit...") or disclaimers ("I cannot access..."). Start directly with
CONTROL:.
Mindset
Moved to references/advanced-patterns.md. Four overlapping finding states, read in dashboard-triage order, reflect current real-world posture rather than the raw status string.
Decision Tree (classification flow)
Apply top-to-bottom. First matching rule wins.
START
│
├─ Compliance block missing? ───────────────► NOT_APPLICABLE (integration finding)
│
├─ RecordState = ARCHIVED? ─────────────────► NOT_APPLICABLE (stale)
│
├─ Workflow.Status = SUPPRESSED + FAILED? ──► WARNING (governance flag)
│
├─ Workflow.Status = RESOLVED + FAILED? ────► WARNING (>48h → FAILED)
│
├─ Compliance.Status = NOT_AVAILABLE?
│ ├─ NO_RESOURCES / DISABLED_CONTROL ────► NOT_APPLICABLE
│ ├─ UNSUPPORTED_INSTANCE_TYPE ─────────► NOT_APPLICABLE
│ ├─ SUPPORTED_SERVICE_NOT_ENABLED ──────► WARNING (Config gap)
│ ├─ ASSUME_ROLE_ERROR / PERMISSION_DENIED► WARNING
│ ├─ CONFIG_RETURNS_UNKNOWN ─────────────► WARNING
│ ├─ ASSESSMENT_FAILED ──────────────────► WARNING
│ ├─ SECURITY_HUB_NOT_ENABLED_IN_REGION ─► WARNING
│ └─ empty / absent ─────────────────────► WARNING (ambiguous)
│
├─ Compliance.Status = FAILED? ─────────────► FAILED
├─ Compliance.Status = WARNING? ────────────► WARNING
├─ Compliance.Status = PASSED? ─────────────► PASSED
│
└─ (unrecognized status) ───────────────────► WARNING + "unrecognized Compliance.Status"
Two rules that cause the most misclassifications — always double-check:
NOT_AVAILABLE+SUPPORTED_SERVICE_NOT_ENABLED= WARNING (not NOT_APPLICABLE — Config being disabled is a monitoring gap).SUPPRESSED+FAILED= WARNING (not PASSED — suppression hides but does not fix).
Process — Classification logic (reference for each step)
Step 0 — Validate input, detect edge cases
Verify the input is an ASFF record. Edge cases to handle before classification:
Malformed JSON: If the finding JSON cannot be parsed, output:
Moved to references/error-handling.md. WARNING emit template for unparseable finding JSON (GetFindings re-fetch, support case for the source integration).
Missing Compliance block: If the Compliance key is absent entirely,
the finding is an integration finding (GuardDuty, Inspector, Macie,
Detective, Firewall Manager) or a custom product finding. It is a security
event, not a control compliance evaluation:
Moved to references/error-handling.md. NOT_APPLICABLE emit template for integration findings (GuardDuty, Inspector, Macie, Detective, Firewall Manager) and custom products.
Malformed ARN fields: If the finding Id or Resources[].Id does not
match the AWS ARN pattern
arn:aws:[a-z0-9-]+:[a-z]{2}-[a-z]+-\d:\d{12}:.*, append:
Moved to references/advanced-patterns.md. DATA_QUALITY flag for finding Id / Resources[].Id not matching the AWS ARN pattern (does not change the verdict).
Field-defaulting rules (apply when a field is absent but non-fatal):
Moved to references/advanced-patterns.md. Defaults: StatusReasons-missing-on-NOT_AVAILABLE → WARNING (never NOT_APPLICABLE), RecordState → ACTIVE, Workflow → NEW, UpdatedAt-absent → STALE_FLAG, Severity → MEDIUM.
Step 1 — RecordState = ARCHIVED → NOT_APPLICABLE (stale)
Security Hub archives findings when the resource was deleted, the control was
disabled, or the finding was superseded by a newer one. An archived finding
represents a historical state. If RecordState = ARCHIVED, emit
NOT_APPLICABLE regardless of Compliance.Status.
Cite "Rule 1: archived — resource deleted or control superseded".
Moved to references/advanced-patterns.md. STALE_FLAG when UpdatedAt is >30 days old even on ACTIVE findings (delegated-admin removal, control re-enable).
Step 2 — Suppressed failure → WARNING (governance flag)
If Workflow.Status = SUPPRESSED AND Compliance.Status = FAILED, emit
WARNING. This is the single most misunderstood Security Hub state.
Suppression hides the finding from the default dashboard view, but the
control is still failing. Suppression is an operational triage decision
(accepted risk, compensating control, false positive), not a remediation.
Cite "Rule 2: suppressed FAILED — governance concern". Always include in the
REASON whether the suppression has a Note:
- If
Workflow.Noteis present, cite it (e.g., "Suppressed by securityops@example.com on 2026-06-15: 'Accepted risk — compensating control via WAF rule arn:aws:wafv2:...'"). - If no
Note, flag: "Suppression has no documented justification — review whether the suppression is still valid." - Security Hub does not natively expire suppressions. A suppression created months ago for a temporary exception may still be hiding a failing control.
Moved to references/advanced-patterns.md. No suppression TTL — validity heuristic by age and Note presence (<30d valid, 30-90d review, >90d stale, no-Note suspect).
Step 3 — Resolved but still FAILED → WARNING (pending re-evaluation)
If Workflow.Status = RESOLVED AND Compliance.Status = FAILED, emit
WARNING. The operator marked the finding resolved (remediation applied),
but Security Hub has not yet re-evaluated the control.
Security Hub re-runs control checks every 12 to 24 hours (the Config rule
evaluation trigger determines the exact interval — see Config-rule evaluation
triggers reference). If the finding was marked RESOLVED more than 48 hours
ago and is still FAILED, the remediation likely failed — escalate to
FAILED and note: "Resolution marked >48h ago but control still FAILED —
remediation likely incomplete. Re-investigate the resource directly."
Step 4 — NOT_AVAILABLE → split by StatusReasons
Compliance.Status = NOT_AVAILABLE means Security Hub could not produce a
definitive PASSED/FAILED. The reason matters — it splits into two verdicts:
NOT_APPLICABLE (control genuinely does not apply):
NO_RESOURCES— zero resources in scope. Example: Redshift.1 with no Redshift clusters.DISABLED_CONTROL— control disabled by the delegated administrator (or via Security Hub Central Configuration).UNSUPPORTED_INSTANCE_TYPE— resource exists but the control does not evaluate this variant.
WARNING (Security Hub tried and failed — needs investigation):
SUPPORTED_SERVICE_NOT_ENABLED— AWS Config not enabled in the region. The control cannot run — a monitoring posture misconfiguration, not a pass.CONFIG_RETURNS_UNKNOWN— Config rule returned UNKNOWN (transient error or permissions issue with the Config service-linked role).ASSESSMENT_FAILED— assessment threw an error. Often transient but may indicate systemic problems (permissions, throttling).ASSUME_ROLE_ERRORorPERMISSION_DENIED— service-linked role lacks permissions to evaluate the resource.SECURITY_HUB_NOT_ENABLED_IN_REGION— Hub disabled in the region.CANNOT_ENABLE_RECORDS— cannot enable Config recording.UNKNOWN_CONTROL_STATUS— undetermined, investigate.- StatusReasons empty or absent → WARNING: "NOT_AVAILABLE with no StatusReasons — investigate manually."
Moved to references/advanced-patterns.md. Complete StatusReason code → verdict → meaning table (NO_RESOURCES through UNKNOWN_CONTROL_STATUS).
Cite "Rule 4: NOT_AVAILABLE — <code>".
Step 5 — Compliance.Status = FAILED → FAILED
RecordState = ACTIVE, Workflow.Status ∈ {NEW, NOTIFIED},
Compliance.Status = FAILED. The control is currently failing and has not
been triaged. Emit FAILED.
Cite "Rule 5: active FAILED — control is non-compliant". Map to the fix-action table for the specific remediation.
Step 6 — Compliance.Status = WARNING → WARNING
Security Hub uses WARNING for partially compliant controls or informational
concerns. Example: a CloudTrail trail exists but log validation is disabled.
Emit WARNING.
Cite "Rule 6: WARNING — partial compliance or informational concern".
Step 7 — Compliance.Status = PASSED → PASSED
Compliance.Status = PASSED AND RecordState = ACTIVE. The resource is
fully compliant. Emit PASSED with REMEDIATION: None required — control is compliant.
Step 8 — Aggregation (multi-finding rollup)
When auditing multiple findings for the same control across resources, or multiple controls across an account, the aggregate verdict is the worst verdict:
FAILED > WARNING > PASSED > NOT_APPLICABLE
Report the count of each verdict in the aggregate:
CONTROL: <ControlId>
AGGREGATE VERDICT: FAILED
BREAKDOWN: 3 FAILED, 1 WARNING, 12 PASSED, 2 NOT_APPLICABLE (across 18 resources)
Severity escalation matrix
Moved to references/advanced-patterns.md. Verdict × severity priority/SLA table, cross-standard severity rule, and blast-radius escalation (aggregate N>10, time-decay >90d, trust-boundary → P0).
Config-rule evaluation triggers (expert reference)
Moved to references/advanced-patterns.md. Config-change vs periodic trigger families per control, 5-30 min vs 6-24h re-evaluation windows, describe-config-rules verification.
Security Hub internals (non-obvious operational knowledge)
Moved to references/advanced-patterns.md. Dedup composite key, BatchUpdateFindings race, Config backlog delays, ingestion latency stages, automation-rule silent suppression, control enablement cost model.
Finding correlation and root-cause grouping
Moved to references/advanced-patterns.md. Root-cause grouping rules (S3 BPA fan-in, account-policy gaps, Config-disabled clusters, cross-standard duplicates).
Coverage assessment heuristic
Moved to references/advanced-patterns.md. coverage_score formula and 90% / 70% operational benchmarks.
Control-to-fix-action mapping
Moved to references/remediation-guidance.md. FSBP (S3, IAM, CloudTrail, Config, EC2, KMS, RDS, Lambda), CIS v1.2-v2.0, and PCI/NIST control → fix action → CLI tables.
Multi-account aggregation (Organizations)
Moved to references/advanced-patterns.md. Delegated-admin aggregation, enrolled-but-not-enabled WARNING, FindingAggregator cross-region propagation, ASSUME_ROLE_ERROR root cause and fix.
Custom controls and non-standard findings
Moved to references/advanced-patterns.md. Custom control remediation lookup chain (Description/RemediationUrl → ASFF Remediation fields → resource-type derivation) and automation-rule detection.
Output format — worked examples
Multi-finding report
CONTROL: S3.2 (S3 Bucket-Level Public Access Prohibition)
VERDICT: FAILED
REASON: Rule 5 (active FAILED) — Compliance.Status is FAILED,
Workflow.Status is NEW, RecordState is ACTIVE. Bucket app-uploads-prod
does not have bucket-level BPA enabled.
SEVERITY: HIGH
REMEDIATION: Enable all 4 bucket-level BPA settings on app-uploads-prod:
aws s3api put-public-access-block --bucket app-uploads-prod
--public-access-block-configuration
BlockPublicAcls=true,IgnorePublicAcls=true,
BlockPublicPolicy=true,RestrictPublicBuckets=true
CONTROL: IAM.7 (IAM Password Policy)
VERDICT: PASSED
REASON: Rule 7 (PASSED) — Compliance.Status is PASSED, RecordState is ACTIVE.
Password policy meets all FSBP requirements.
SEVERITY: MEDIUM
REMEDIATION: None required — control is compliant.
CONTROL: Redshift.1 (Redshift Clusters Public Access)
VERDICT: NOT_APPLICABLE
REASON: Rule 4 (NOT_AVAILABLE — NO_RESOURCES) — Compliance.Status is
NOT_AVAILABLE with StatusReasons code NO_RESOURCES. No Redshift clusters
exist in this account/region.
SEVERITY: MEDIUM
REMEDIATION: None required — control does not apply.
CONTROL: Config.1 (AWS Config Enabled)
VERDICT: WARNING
REASON: Rule 4 (NOT_AVAILABLE — SUPPORTED_SERVICE_NOT_ENABLED) —
Compliance.Status is NOT_AVAILABLE with StatusReasons code
SUPPORTED_SERVICE_NOT_ENABLED. AWS Config is not enabled in eu-west-2,
so this control cannot evaluate. This is a monitoring gap, not a pass.
SEVERITY: MEDIUM
REMEDIATION: Enable AWS Config in eu-west-2:
aws configservice put-configuration-recorder
--configuration-recorder name=default,roleARN=<role>
--recording-group allSupported=true,includeGlobalResourceTypes=true
NEVER (anti-patterns)
-
NEVER pass through
Compliance.Status = NOT_AVAILABLEas NOT_APPLICABLE without checkingStatusReasons.NO_RESOURCESmeans the control does not apply (NOT_APPLICABLE), butSUPPORTED_SERVICE_NOT_ENABLEDmeans Config is off — a monitoring posture gap that is a WARNING. Treating the latter as NOT_APPLICABLE hides a misconfiguration that makes the entire compliance posture unreliable. -
NEVER treat a SUPPRESSED finding as PASSED.
Workflow.Status = SUPPRESSEDmeans someone hid the finding from the default view. IfCompliance.Status = FAILED, the control is still failing — the verdict is WARNING. Suppression is a triage decision, not a remediation. -
NEVER treat
Workflow.Status = RESOLVEDas evidence that the resource is now compliant. RESOLVED is an operator assertion. Security Hub re-evaluates on its own schedule. If stillFAILEDafter 48h (or 1h for config-change controls), the fix failed. -
NEVER classify an ARCHIVED finding as FAILED.
RecordState = ARCHIVEDmeans the resource was deleted or the control superseded. The verdict is NOT_APPLICABLE. -
NEVER ignore
FirstObservedAt/UpdatedAttimestamps. AFAILEDfinding open for 180 days is a different operational concern than one opened yesterday. Age compounds risk — escalate for findings > 90 days old. -
NEVER classify an integration finding (GuardDuty, Inspector, Macie, Detective, Firewall Manager) as a control compliance verdict. These lack a
Complianceblock — they are security events. Route them to the appropriate detector skill. -
NEVER assume a single region's findings represent global compliance. Security Hub is regional. A finding for
us-east-1says nothing abouteu-west-1. Always check theRegionfield. In multi-region accounts, the delegated administrator must aggregate viaFindingAggregator. -
NEVER recommend disabling a Security Hub control as the fix for a FAILED finding. Disabling removes visibility — it trades FAILED for NOT_APPLICABLE without fixing the resource. The only legitimate reason to disable is a confirmed false-positive.
-
NEVER aggregate findings by severity alone without considering blast radius. A
MEDIUMFAILED on an IAM role with cross-account trust is more dangerous than tenHIGHfindings on isolated dev resources. -
NEVER assume
Severity.Labelis the final priority. The aggregate (N > 10) and time-decay (> 90 days) escalation rules override the label — aMEDIUMFAILED on 50 resources for 6 months is effectively P1. -
NEVER batch-update finding workflow status to RESOLVED without confirming the resource is actually fixed. If the resource is still non-compliant, Security Hub re-opens the finding on the next cycle.
-
NEVER trust ARN fields blindly. A finding
Resources[].Idthat does not match the AWS ARN pattern (arn:aws:[service]:[region]:[account]:[resource]) indicates either a malformed ASFF payload or a non-AWS resource incorrectly ingested. Flag the data-quality issue and verify the finding source before acting on the remediation. Malformed ARNs cause downstream failures in any CLI command that takes--resource-arnor--bucketparameters. -
NEVER assume cross-account Security Hub findings are actionable from the administrator account. The delegated administrator can SEE findings from member accounts but may not have permission to MODIFY the underlying resources. A control backed by a Config rule that assumes
AWSServiceRoleForSecurityHubin the member account will showASSUME_ROLE_ERRORif that role's trust policy is broken or the member account was enrolled without enabling Security Hub. This is a WARNING — the role trust must be fixed before the control can evaluate. -
NEVER call
BatchUpdateFindingswith more than 100 finding identifiers. The API rejects batches over 100 withInvalidInputException. Chunk into groups of 100 or fewer. -
NEVER apply remediation commands in a tight loop without exponential backoff. Config rule evaluations and Security Hub API calls share a rate limit. A burst across 50 resources can trigger
ThrottlingExceptionand leave some unremediated silently. -
NEVER attempt to update
Compliance.StatusviaBatchUpdateFindings— it is a system-managed field. OnlyWorkflow.Status,RecordState,Note, andSeverityare user-updatable. Trying to setCompliance.StatusreturnsInvalidInputException.
API and CLI constraints
Moved to references/error-handling.md. Pagination and batch limits (100 findings, 4 KB Note), throttling/backoff, eventual consistency, CLI failure recovery (InvalidInputException, AccessDenied, NoSuchBucket).
Pre-flight safety checks (run before any remediation)
Moved to references/diagnostic-commands.md. Dry-run mode for batch remediation, resource-existence probes per service, rollback state capture, region/delegated-admin verification, trust-boundary incident handling, permission-fallback escalation.
Remediation guidance
Moved to references/remediation-guidance.md. FAILED playbook (map → pre-flight → fix → RESOLVED → await re-evaluation), WARNING playbook (suppression review/unsuppress, resolved-pending, Config gap), NOT_APPLICABLE disposition.
Recent AWS features (2024-2026)
Moved to references/advanced-patterns.md. Central configuration, Automated Security Response on AWS, control lifecycle updates, Audit Manager integration.
References (load on demand)
- advanced-patterns — Activation triggers, Mindset, Step 0-4 detail prose, severity escalation matrix, Config-rule triggers, Security Hub internals, correlation, coverage heuristic, multi-account aggregation, custom controls, recent AWS features
- diagnostic-commands — pre-flight safety checks (dry-run, existence probes, rollback capture)
- error-handling — malformed-input emit blocks and API/CLI constraints with failure recovery
- remediation-guidance — control-to-fix-action mapping tables and per-verdict remediation playbooks
Domain
AWS CloudOps / Security Hub Compliance & Posture Management.
AWS documentation
- AWS Security Hub User Guide — https://docs.aws.amazon.com/securityhub/latest/userguide/what-is-securityhub.html
- Security Hub Security — https://docs.aws.amazon.com/securityhub/latest/userguide/security.html
- Security Hub API Reference — https://docs.aws.amazon.com/securityhub/1.0/APIReference/
- Security Hub CLI Reference — https://docs.aws.amazon.com/cli/latest/reference/securityhub/
- Central configuration — https://docs.aws.amazon.com/securityhub/latest/userguide/central-configuration.html
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.