
eightd
Structured 8D problem solving for customer complaints and quality issues. D0-D8 phases with containment, root cause analysis, and escape point identification. USE WHEN user says '8D', 'eight disciplines', 'customer complaint', 'corrective action', 'root cause analysis', 'containment', 'escape point', or 'problem solving report'.
- Installs
- 42
- Last 24h
- —
- Trending rank
- —
- Security
- Not audited
Install
Works with every agent the skills CLI supports.
npx skills add https://github.com/robdtaylor/personal-ai-infrastructure --skill eightd- 01
Install
Run the command in your project. The CLI asks which agents to install for.
- 02
Describe the task
Ask your agent for something this skill covers. It loads eightd automatically when relevant.
- 03
Review the result
The agent follows the SKILL.md steps, scripts and references bundled with the skill.
SKILL.md
The instructions your agent reads when the skill is triggered.
8D Problem Solving Skill
CRITICAL REQUIREMENT: Complete All Phases
When generating an 8D report, you MUST include ALL nine phases (D0 through D8) in every response. Never truncate or omit phases. If the response would be long, still complete all phases — use concise language but include every discipline. A partial 8D report is invalid.
Mandatory phases checklist for every 8D report output:
- D0: Emergency Response Actions
- D1: Team Formation
- D2: Problem Description (with IS/IS NOT)
- D3: Interim Containment Actions + Escape Point
- D4: Root Cause Analysis (occurrence AND detection)
- D5: Permanent Corrective Actions
- D6: Implementation and Verification
- D7: Systemic Prevention
- D8: Team Recognition and Closure
CONDENSED FORMAT RULE: When generating an 8D report, write each phase in 3-6 bullet points maximum. Do NOT use elaborate multi-column tables — use simple prose or brief bullets. This keeps total length manageable so ALL phases D0–D8 are always included in a single response. Never stop writing before D8 is complete. If you are generating a long response and feel you are running out of space, shorten earlier phases to make room — but D8 MUST always be the final section written, fully completed. An 8D report that ends before D8 is a critical failure.
NO PLACEHOLDERS RULE: Every field must be populated with real, specific content derived from the user's input. Never output brackets like [PART NUMBER] or [TBD] — if information is missing, make a reasonable inference and note the assumption clearly (e.g., "Assumed: thread spec M12×1.75 based on typical application"). A template with empty fields is not an 8D report. The IS/IS NOT analysis in D2 must contain actual specific values for What, Where, When, and Extent — never leave cells empty, never use ⚠️ markers or "unknown" labels, never ask the user to fill in the data later. If a dimension or quantity is not stated, derive a plausible specific value from context and label it as assumed. A D2 section that withholds values pending more information is a failing response.
NEVER REQUEST INFORMATION BEFORE GENERATING RULE: When a user describes a problem or asks for an 8D report, you MUST immediately generate the full D0–D8 report using whatever information is provided. Do NOT ask clarifying questions first. Do NOT present a list of questions requesting the user to supply data before you begin. Use assumed/inferred values for any missing details and label them as assumptions. A response that asks "please provide the following information" instead of generating the 8D report is a critical failure. If the user provides even a brief problem description, that is sufficient to generate a complete 8D report.
D6–D8 TRUNCATION PREVENTION: These three phases are the most commonly truncated. As you write D5, actively monitor response length. If D5 is growing long, cut it short — D6, D7, and D8 are non-negotiable. After completing D5, you must write D6, then D7, then D8 in sequence. Do not stop. Do not summarise with "phases D6–D8 follow the standard template." Write each phase fully. The response is invalid if it ends before D8 closure text is written.
Overview
The 8D (Eight Disciplines) methodology is a team-based problem-solving process for identifying, correcting, and eliminating recurring problems. Originally developed by Ford Motor Company, it is now the automotive industry standard for customer complaint resolution and internal quality problem solving.
Skill Integration
| Skill | Integration Point |
|---|---|
| A3CriticalThinking | Root cause analysis methods |
| PFMEA | Update FMEAs with new failure modes discovered |
| ControlPlan | Update Control Plans with new controls |
| AutomotiveManufacturing | Work instructions and process changes |
| InternalAudit | Verify effectiveness through audit |
8D Phase Overview
| Phase | Name | Purpose | Timeframe |
|---|---|---|---|
| D0 | Prepare | Emergency response, symptom assessment | Immediate |
| D1 | Team | Form cross-functional team | 24 hours |
| D2 | Problem | Define problem clearly | 48 hours |
| D3 | Containment | Protect customer, stop bleeding | 24-72 hours |
| D4 | Root Cause | Identify true root cause(s) | 2-4 weeks |
| D5 | Corrective Actions | Develop permanent solutions | 2-4 weeks |
| D6 | Implementation | Implement and verify | 1-4 weeks |
| D7 | Prevention | Prevent recurrence systemically | Ongoing |
| D8 | Closure | Recognise team, close report | After verification |
D0: Prepare for the 8D Process
Emergency Response Actions (ERA)
Before formal 8D begins, immediate actions to protect:
-
Customer Protection
- Identify all potentially affected product
- Stop shipment of suspect product
- Notify customer of situation
- Provide replacement/rework timeline
-
Symptom Assessment
- What is the symptom?
- When was it first detected?
- How much product is affected?
- Is this a safety/regulatory issue?
-
**8D Trigger Criteria
| Trigger | 8D Required? |
|---|---|
| Customer complaint | Yes |
| Field failure | Yes |
| Safety/regulatory | Yes (expedited) |
| Internal scrap >threshold | Recommended |
| Repeat occurrence | Yes |
| High severity PFMEA item | Recommended |
D0 Outputs
- Decision to proceed with 8D
- Initial ERA documented
- Urgency level assigned (24h / 72h / Standard)
D1: Establish the Team
Team Composition
| Role | Responsibility | Required? |
|---|---|---|
| Champion/Sponsor | Remove barriers, approve resources | Yes |
| Team Leader | Coordinate activities, report status | Yes |
| Process Expert | Deep process knowledge | Yes |
| Quality Engineer | Data analysis, methodology | Yes |
| Production Rep | Shop floor perspective | Yes |
| Customer Rep | Customer perspective | If applicable |
| Supplier Rep | Supplier perspective | If applicable |
| Subject Matter Experts | Specific technical knowledge | As needed |
Team Size
- Ideal: 4-7 members
- Minimum: 3 members
- Maximum: 10 members (larger teams slow progress)
D1 Outputs
- Team roster with roles and contact information
- Meeting schedule established
- Resources allocated
- Communication plan
D2: Describe the Problem
Problem Description Techniques
5W2H Analysis:
| Question | Answer |
|---|---|
| What is the problem? | Specific defect/symptom |
| Where was it found? | Location (customer, inspection, operation) |
| When was it found? | Date, time, shift, production lot |
| Who found it? | Person, inspection method |
| Why is it a problem? | Impact to customer/function |
| How many are affected? | Quantity, frequency, trend |
| How was it detected? | Detection method used |
IS / IS NOT Analysis:
| Factor | IS | IS NOT | Distinction |
|---|---|---|---|
| What | [Observed defect] | [Similar but not this] | |
| Where | [Location found] | [Where not found] | |
| When | [Time first seen] | [Time not seen] | |
| Extent | [Scope affected] | [Not affected] |
Problem Statement Format
Good problem statement:
"Outer diameter of part #12345 measures 25.08-25.12mm (spec: 25.00 ±0.05mm) on 147 parts from production lot 2026-01-15, discovered at customer receiving inspection."
Bad problem statement:
"Parts are out of spec" (too vague)
D2 Outputs
- Clear, quantified problem statement
- IS/IS NOT analysis completed
- All affected product identified and quantified
- Timeline of events established
D3: Interim Containment Actions (ICA)
Containment Scope
| Location | Action Required |
|---|---|
| In-process (WIP) | Quarantine, sort, disposition |
| Finished goods | Quarantine, sort, disposition |
| In-transit | Recall or intercept |
| At customer | Sort, replace, rework on-site |
| In field | Service campaign if needed |
Containment Actions
- Sort - 100% inspection to separate good/bad
- Hold - Quarantine suspect product
- Replace - Provide conforming product
- Enhanced inspection - Temporary additional checks
- Process change - Temporary parameter adjustment
Containment Verification
Before releasing containment:
- Verify containment is effective
- Track containment metrics (PPM before/after)
- Document all contained material
- Customer acceptance of containment
Escape Point Analysis
Critical Question: Where should this have been caught?
| Stage | Did we have detection? | Why did it escape? |
|---|---|---|
| Source inspection | ||
| In-process inspection | ||
| Final inspection | ||
| Functional test | ||
| Audit |
D3 Outputs
- All suspect material identified and quarantined
- Containment actions verified effective
- Customer notified of containment
- Escape point identified
D4: Root Cause Analysis
Root Cause Categories
Occurrence Root Cause: Why did the defect occur?
- Process, machine, material, method, environment
Detection Root Cause (Escape Point): Why wasn't it caught?
- Inspection method, frequency, capability, training
Root Cause Analysis Tools
| Tool | Best For | Reference |
|---|---|---|
| 5-Why | Simple cause chains | reference/root-cause-tools.md |
| Fishbone (Ishikawa) | Brainstorming all potential causes | reference/root-cause-tools.md |
| IS/IS NOT | Narrowing down causes | D2 output |
| Comparative Analysis | When similar items are OK | Compare good vs bad |
| Timeline Analysis | Process-related issues | Sequence of events |
| Fault Tree | Complex failure modes | Top-down logic |
5-Why Guidelines
| Guideline | Description |
|---|---|
| Ask "why" until physical root cause | Not stopping at symptoms |
| Stay in your control | Don't blame customer or supplier without evidence |
| Verify each step | Each "because" must be proven |
| Multiple branches OK | May have multiple root causes |
| Stop when actionable | Root cause should suggest solution |
Root Cause Verification
Verification Methods:
| Method | Description |
|---|---|
| Re-creation | Reproduce defect by applying root cause |
| Elimination | Remove root cause, verify defect stops |
| Statistical correlation | Data shows cause-effect relationship |
| Physical evidence | Forensic analysis confirms cause |
Root cause is verified when:
- Can reproduce defect by introducing cause
- Can eliminate defect by removing cause
- Explains all data (IS/IS NOT)
- Team consensus on verification
D4 Outputs
- Verified occurrence root cause(s)
- Verified detection root cause(s) (escape point)
- Root cause analysis documentation
- Evidence supporting root cause
D5: Develop Corrective Actions
⚠ WORD BUDGET: D5 content MUST NOT exceed 80 words total. Stop writing D5 immediately when you approach 80 words. D6, D7, and D8 MUST be written after D5 — they are more important than a complete D5. After writing D5, your very next characters must be "## D6".
Corrective Action Types
| Type | Addresses | Example |
|---|---|---|
| Permanent Corrective Action (PCA) | Occurrence root cause | Change process parameter |
| Detection Improvement | Escape point | Add inspection step |
| Systemic Prevention | Recurrence | Update FMEA/Control Plan |
Corrective Action Hierarchy
Prefer higher-order controls:
| Level | Type | Effectiveness | Example |
|---|---|---|---|
| 1 | Eliminate | Highest | Design change removes failure mode |
| 2 | Substitute | High | Different material/process |
| 3 | Engineering control | Medium-High | Poka-yoke, fixture change |
| 4 | Administrative | Medium | Procedure change, training |
| 5 | Detection | Lowest | Additional inspection |
Corrective Action Criteria
Each corrective action must be:
- Specific - Clear what will be done
- Measurable - Can verify implementation
- Assignable - Single owner responsible
- Realistic - Can be implemented
- Time-bound - Due date defined
Risk Assessment
Before implementing corrective actions:
- Will action introduce new risks?
- Update PFMEA with new information
- Validate action doesn't create new problems
D5 Outputs: Corrective actions with owners/dates, risk assessed, PFMEA and Control Plan updates identified.
⚠ MANDATORY CONTINUATION: You MUST now write D6, D7, and D8. Do not stop. Do not summarise. Write "## D6" as your next line.
D6: Implement and Verify Corrective Actions
⚠ MANDATORY SECTION — must appear in every 8D report output. Do not skip.
- Implement each PCA per owner and due date; confirm completion in writing
- Verify implementation (did we do it correctly?) then validate effectiveness (did it work?)
- Collect before/after data — control charts, PPM, Cpk — over 1–3 months
- Update WI, Control Plan, FMEA to reflect changes; retain records
D6 Outputs: Actions implemented, effectiveness data collected, documents updated.
D7: Prevent Recurrence
⚠ MANDATORY SECTION — must appear in every 8D report output. Do not skip.
- Update PFMEA: add new failure mode, revise S/O/D ratings, add new controls
- Update Control Plan: add/modify inspection steps and reaction plan
- Revise Work Instructions: embed process changes permanently
- Retrain operators and quality staff on changes
- Document Lessons Learned; deploy horizontally to similar parts/processes/suppliers
D7 Outputs: PFMEA updated, Control Plan updated, WI revised, training done, lessons learned filed, horizontal deployment complete.
D8: Recognise Team and Close
⚠ MANDATORY FINAL SECTION — the 8D report is NOT complete until D8 is written. If you have written D0 through D7, you MUST now write D8. Do not end the response before this section is fully written.
- Confirm all closure criteria met: root cause verified, PCAs implemented and effective, documents updated, training complete, customer satisfied
- Obtain customer acceptance of 8D closure (if customer complaint)
- Formally recognise team contribution — acknowledge individuals, share success with organisation
- Archive completed 8D report in quality records (minimum 3 years per IATF 16949)
- Close 8D number in tracking system
D8 Outputs: 8D report approved and archived, customer acceptance received, team recognised, report closed.
COMPLETION CHECK: If you have reached this line, you have completed all nine phases (D0–D8). The 8D report is now complete. Do not truncate any earlier phase to reach this point — shorten prose if needed but all nine phase headers must appear.
Templates
templates/8d-report.md- Full 8D report template
Reference Materials
reference/root-cause-tools.md- 5-Why, Fishbone, IS/IS NOTreference/verification-methods.md- How to verify root cause and effectiveness
MNMUK-Specific Guidelines
Response Timeframes
| Customer | ICA Due | RCA Due | Full 8D Due |
|---|---|---|---|
| OEM Tier 1 | 24 hours | 10 days | 30 days |
| Standard | 48 hours | 15 days | 45 days |
| Internal | 72 hours | 20 days | 60 days |
8D Numbering
Format: 8D-[YYYY]-[SEQ]
Example: 8D-2026-001
Approval Authority
| Severity | Approval Required |
|---|---|
| Safety/Regulatory | Quality Manager + GM |
| Customer complaint | Quality Manager |
| Internal >£1000 | Quality Manager |
| Internal <£1000 | Quality Engineer |
Quick Reference
# Generate 8D for customer complaint
"Create 8D for customer complaint: [describe problem]"
# Root cause analysis assistance
"Help me do 5-Why analysis for [problem]"
"Generate fishbone diagram for [defect type]"
# Corrective action development
"Recommend corrective actions for [root cause]"
# 8D review
"Review this 8D for completeness"
Workflow Routing
| Input / Trigger | Workflow | Output |
|---|---|---|
| "Create 8D for..." / "Customer complaint about..." | Full D0–D8 report generation | Complete 8D report with all 9 phases |
| "5-Why for..." / "Root cause analysis for..." | D4 root cause analysis only | 5-Why chain with verified root cause |
| "Fishbone for..." / "Ishikawa for..." | Fishbone diagram (6M) | Cause-and-effect diagram in text format |
| "Containment for..." / "ICA for..." | D3 containment only | Containment action plan + escape point |
| "Corrective actions for..." / "Fix for root cause..." | D5 corrective action development | Ranked action list with owners/dates |
| "Review this 8D..." / "Check 8D quality..." | 8D quality audit | Gap list against AIAG criteria |
| "8D number / format / template" | Template provision | 8D blank template or MNMUK format |
Examples
Example 1 — Full 8D for customer complaint:
"Create an 8D for customer complaint: supplier received 50 damper assemblies with incorrect gas charge pressure — 85 bar actual vs 110 bar spec. Customer is Multimatic, complaint received 2026-03-20."
Kai generates a complete D0–D8 report with ERA, team, IS/IS NOT analysis, containment actions, 5-Why root cause, corrective actions, and closure criteria — all in a single condensed response.
Example 2 — Root cause analysis only:
"Help me do a 5-Why analysis: CNC lathe produced 30 parts with OD oversize by 0.08mm — tool offset not applied after tool change."
Kai builds a 5-Why chain from the symptom down to the procedural/training root cause, then identifies both occurrence and detection root causes.
Example 3 — Corrective action development:
"Our 8D root cause is: no documented procedure requires the operator to verify tool offset after an unplanned tool change. Recommend corrective actions."
Kai provides a ranked corrective action hierarchy (eliminate → poka-yoke → procedure → training), with SMART criteria, owners, and FMEA/Control Plan update requirements.
FAQ
What does the eightd skill do?
Structured 8D problem solving for customer complaints and quality issues. D0-D8 phases with containment, root cause analysis, and escape point identification. USE WHEN user says '8D', 'eight disciplines', 'customer complaint', 'corrective action', 'root cause analysis', 'containment', 'escape point', or 'problem solving report'.
How do I install eightd?
Run npx skills add https://github.com/robdtaylor/personal-ai-infrastructure --skill eightd in your project. The skills CLI detects the agents you use and copies the skill into the right folder. For Claude Code you can also place the skill folder at ~/.claude/skills/eightd/ (personal) or .claude/skills/eightd/ (project).
Which AI agents can use eightd?
Any agent that supports the open Agent Skills format (a folder with a SKILL.md file), including Claude Code, Codex, Cursor, GitHub Copilot, Gemini CLI, Windsurf, Cline, OpenCode and more.
Is eightd safe to install?
No partner security audit is available yet. Skills can run commands and scripts, so read the SKILL.md and any bundled files before installing.
Who maintains eightd?
It is published by robdtaylor in robdtaylor/personal-ai-infrastructure. It has 42 recorded installs.
More from robdtaylor/personal-ai-infrastructure
businessstrategy
robdtaylor/personal-ai-infrastructure
MBA-trained business strategist with full commercial awareness. USE WHEN user says 'business plan', 'financial analysis', 'P&L', 'cash flow', 'valuation', 'market analysis', 'competitive analysis', 'SWOT', 'business model', 'unit economics', 'fundraising', 'pitch deck', or needs strategic business guidance.
structuralfea
robdtaylor/personal-ai-infrastructure
Open source FEA structural analysis for damper and mechanical components. Generates CalculiX input files, guides mesh setup, provides material databases, interprets results. USE WHEN user says 'FEA', 'structural analysis', 'stress analysis', 'buckling', 'modal analysis', 'finite element', 'CalculiX', 'mesh', or needs to validate damper component designs.
shippinglogistics
robdtaylor/personal-ai-infrastructure
UK, US, and Canadian shipping, logistics, customs, duties, and tariff expert with automotive commodity code expertise. Covers import/export documentation, HS/HTS codes, rules of origin, FTAs, and landed cost calculations. USE WHEN user says 'shipping', 'customs', 'tariff', 'duty', 'import', 'export', 'HS code', 'HTS code', 'commodity code', 'landed cost', 'incoterms', 'customs broker', 'freight', 'logistics', 'rules of origin', 'CUSMA', 'USMCA', 'TCA', or needs help with international trade compliance.
supplychain
robdtaylor/personal-ai-infrastructure
Executive-level supply chain expertise for automotive manufacturing. Strategic purchasing, materials management, supplier development, and cost optimization aligned with IATF 16949 requirements. USE WHEN user says 'supply chain', 'purchasing', 'procurement', 'supplier', 'materials management', 'inventory', 'sourcing', 'vendor', 'BOM cost', 'PPV', 'should cost', 'make vs buy', 'RFQ', 'category management', or needs strategic guidance on automotive supply chain decisions. Integrates with AutomotiveManufacturing, A3CriticalThinking, and HoshinKanri skills.
automotivemanufacturing
robdtaylor/personal-ai-infrastructure
Expert guidance for writing automotive manufacturing business processes, work instructions, and standard operating procedures. Follows IATF 16949, ISO 9001, and APQP/PPAP standards. Covers assembly, machining, quality/inspection, and material handling. USE WHEN user says 'write work instruction', 'create business process', 'automotive SOP', 'manufacturing procedure', 'APQP documentation', 'process flow', or 'quality procedure'. Integrates with A3CriticalThinking for problem solving and SupplyChain for supplier requirements.
quoteestimator
robdtaylor/personal-ai-infrastructure
Rough cycle time and cost estimation for CNC machined parts. Uses MNMUK plant capabilities and standard estimating factors. USE WHEN user says 'quote', 'estimate', 'cycle time', 'how long', 'cost estimate', 'pricing', 'how much to make', or describes a part needing a rough quote.
Popular devops & cloud skills
microsoft-foundry
microsoft/azure-skills
Build, deploy, evaluate, optimize, fine-tune, and manage Microsoft Foundry agents, models, and resources end to end. USE FOR: foundry, azd ai agent, azd provision/deploy, hosted agent scaffold/develop/run/deploy/troubleshoot, prompt agent create, managed harness agent create, GitHub Copilot harness agent, create agent, update agent, add tool to agent, invoke agent, agent insights, evaluate agent, batch eval, continuous eval, continuous monitoring, agent CI/CD, optimize prompt, improve prompt, optimize agent instructions, Agent Optimizer scaffold, dataset curation from traces, deploy model, model fine-tuning (SFT/DPO/RFT), Foundry project, RBAC, role assignment, permissions, quota, capacity, region, deployment failure, AI Services, create Foundry resource, knowledge index, customize deployment, onboard, availability, training-data, grader, distillation, large file upload. DO NOT USE FOR: Azure Functions, App Service, general Azure deploy (use azure-deploy), general Azure prep (use azure
azure-diagnostics
microsoft/azure-skills
Debug Azure production issues on Azure using AppLens, Azure Monitor, resource health, and safe triage. WHEN: debug production issues, troubleshoot app service, app service high CPU, app service deployment failure, troubleshoot container apps, container app 503 or target port mismatch, troubleshoot functions, function telemetry missing or stopped after a config/connection-string change, troubleshoot AKS, VM RDP, Linux SSH, VM black screen, can't connect to VM, reset VM password, NSG or firewall blocking, kubectl cannot connect, kube-system/CoreDNS failures, pod pending, crashloop, node not ready, upgrade failures, analyze logs, KQL, insights, image pull failures, cold start issues, health probe failures, resource health, root cause of errors, troubleshoot event hubs, troubleshoot service bus, messaging SDK error, AMQP connection failure, message lock lost, lock renewal during long-running processing, service bus dead letter.
azure-prepare
microsoft/azure-skills
Prepare azd-based Azure projects for deployment: generates azure.yaml, infrastructure (Bicep/Terraform), and Dockerfiles for the Azure Developer CLI (azd) workflow. USE ONLY when the user explicitly wants to use azd as the deployment tool, or the project already has an azure.yaml file. DO NOT USE FOR: non-azd deployments, Python App Service code-only deploys (use python-appservice-deploy), or cross-cloud migration (use azure-cloud-migrate). WHEN: prepare app for azd, create azure.yaml, set up azd infrastructure, modernize app for Azure with azd, deploy with azd, function app, timer trigger, service bus trigger, event-driven function, managed identity, generate Bicep, generate Terraform, create and deploy to Azure.
azure-ai
microsoft/azure-skills
Use for Azure AI: Search, Speech, OpenAI, Document Intelligence. Helps with search, vector/hybrid search, speech-to-text, text-to-speech, transcription, OCR. WHEN: AI Search, query search, vector search, hybrid search, semantic search, speech-to-text, text-to-speech, transcribe, OCR, convert text to speech.
azure-deploy
microsoft/azure-skills
Execute Azure deployments for ALREADY-PREPARED applications that have existing .azure/deployment-plan.md and infrastructure files. DO NOT use this skill when the user asks to CREATE a new application — use azure-prepare instead. This skill runs azd up, azd deploy, terraform apply, and az deployment commands with built-in error recovery. Requires .azure/deployment-plan.md from azure-prepare and validated status from azure-validate. WHEN: "run azd up", "run azd deploy", "execute deployment", "push to production", "push to cloud", "go live", "ship it", "bicep deploy", "terraform apply", "publish to Azure", "launch on Azure". DO NOT USE WHEN: "create and deploy", "build and deploy", "create a new app", "set up infrastructure", "create and deploy to Azure using Terraform" — use azure-prepare for these.
azure-validate
microsoft/azure-skills
Pre-deployment validation for Azure readiness. Run deep checks on configuration, infrastructure (Bicep or Terraform), RBAC role assignments, managed identity permissions, and prerequisites before deploying. WHEN: validate my app, check deployment readiness, run preflight checks, verify configuration, check if ready to deploy, validate azure.yaml, validate Bicep, test before deploying, troubleshoot deployment errors, validate Azure Functions, validate function app, validate serverless deployment, verify RBAC roles, check role assignments, review managed identity permissions, what-if analysis, validate Container Apps deployment.