Synthesize all quality assessments (risk, test-design, traceability, NFR) into evidence-based gate decision (PASS/CONCERNS/FAIL/WAIVED) with comprehensive rationale. Generates both YAML (CI/CD) and Markdown (human review) reports with action items. Use during final quality review to make go/no-go deployment decisions based on comprehensive quality evidence.
Standardmäßig ist der Prompt ausgewählt, der zuerst die Quelle prüft. Sie können zu einem direkten Befehl wechseln oder eine lokale Kopie herunterladen.
Quelldateien prüfen
Lesen Sie SKILL.md und alle von SkillsMP angezeigten Begleitdateien, bevor Sie sich für eine Installation entscheiden.
Mit Codex oder Claude installieren Kopieren Sie diesen Prompt, fügen Sie ihn in Codex, Claude oder einen anderen Assistant ein und lassen Sie die Skill-Seite prüfen und installieren.
Ein direkter Befehl überspringt den Prüf-Prompt. Prüfen Sie die Quelle, bevor Sie ihn ausführen.
Synthesize all quality assessments (risk, test-design, traceability, NFR) into evidence-based gate decision (PASS/CONCERNS/FAIL/WAIVED) with comprehensive rationale. Generates both YAML (CI/CD) and Markdown (human review) reports with action items. Use during final quality review to make go/no-go deployment decisions based on comprehensive quality evidence.
version
2
category
Quality
acceptance
{"all_assessments_synthesized":"All available quality assessments (risk profile, test design, traceability, NFR) loaded and synthesized into dimension scores","decision_made":"Evidence-based gate decision made (PASS/CONCERNS/FAIL/WAIVED) with clear rationale and criteria evaluation","reports_generated":"Both YAML (CI/CD automation) and Markdown (human review) reports generated with complete audit trail","action_items_identified":"Action items identified for CONCERNS/FAIL decisions with priorities and recommendations"}
inputs
{"task_id":{"required":true,"description":"Task identifier for gate assessment (e.g., 'task-007')"},"task_file":{"required":true,"description":"Path to task specification file"},"config_file":{"required":false,"description":"Path to project configuration (defaults to .claude/config.yaml)"},"require_all_assessments":{"required":false,"description":"If true, all 4 assessments required; if false, minimum 1 required (defaults to false)"}}
outputs
{"gate_decision":{"description":"Final gate decision (PASS/CONCERNS/FAIL/WAIVED)"},"overall_quality_score":{"description":"Overall quality score (0-100)"},"dimension_scores":{"description":"Scores for all 6 quality dimensions"},"critical_issues_count":{"description":"Number of critical issues blocking gate"},"action_items_count":{"description":"Number of action items for remediation"},"yaml_report_path":{"description":"Path to YAML report (for CI/CD automation)"},"markdown_report_path":{"description":"Path to Markdown report (for human review)"}}
The quality-gate skill performs the final quality gate assessment by synthesizing all previous quality evaluations and making an evidence-based decision on whether the implementation is ready to proceed (merge, deploy, release). This skill integrates risk management, test coverage, requirements traceability, and non-functional requirements into a comprehensive quality score across 6 dimensions, then applies configurable criteria to make a gate decision with clear rationale.
Unlike individual quality assessments that evaluate specific aspects (risks, tests, NFRs), the quality gate provides a holistic view of overall quality by weighting and combining all assessments. The gate produces two outputs: a YAML report for CI/CD automation and a markdown report for human review, enabling both automated and manual quality processes. Decisions include PASS (all criteria met), CONCERNS (some issues but can proceed), FAIL (critical issues block proceeding), or WAIVED (issues acknowledged and waived with justification).
The assessment evaluates 6 quality dimensions with configurable weights: Risk Management (25%), Test Coverage (20%), Traceability (20%), NFR (25%), Implementation Quality (5%), and Compliance (5%). Each dimension is scored independently, then combined into an overall quality score. Critical gap rules (e.g., no critical security gaps) and dimension minimums (e.g., traceability ≥65%) are applied to determine if the gate passes.
When to Use This Skill
This skill should be used when:
Final quality decision is needed before merge/deploy/release
All quality assessments need to be synthesized into overall quality score
Audit trail for compliance documentation is required
Quality checks need to be integrated into CI/CD pipeline
Quality waivers need to be tracked with justification
Determination of whether implementation meets quality standards is needed
This skill is particularly valuable:
After all quality assessments complete (risk, test-design, trace, nfr)
As final checkpoint before code merge or deployment
For regulatory compliance requiring audit trail
When automated quality gates are part of CI/CD
For tracking quality trends over time (score history)
This skill should NOT be used when:
Quality assessments haven't been performed yet (run assessments first)
Task is in early planning stage (no implementation to assess)
You only need specific assessment (use individual assessment skills instead)
Prerequisites
Before running quality-gate, ensure you have:
Task specification file with implementation details
Project configuration (.claude/config.yaml) with gate criteria
At least one quality assessment completed:
Minimum: Traceability OR NFR (functional or quality attributes)
Recommended: All 4 assessments (risk, test-design, traceability, nfr)
Waiver requests documented (if any gaps require waiving)
Optional but recommended:
All 4 quality assessments for comprehensive gate
Gate criteria configured in project config
CI/CD pipeline configured to consume YAML output
Sequential Quality Gate Process
This skill executes through 9 sequential steps (Step 0 + Steps 1-8). Each step must complete successfully before proceeding. The process systematically synthesizes all quality dimensions, calculates scores, applies criteria, and makes the final gate decision.
Step 0: Load Configuration and Assessments
Purpose: Load project configuration, task specification, and all available quality assessment reports. Determine which assessments are available and verify minimum requirements are met.
Actions:
Load project configuration from .claude/config.yaml (gate criteria, dimension weights, thresholds)
See:references/templates.md#step-0-configuration-loading-output for complete format
See:references/gate-dimensions.md for dimension definitions
Step 1: Synthesize Risk Management Dimension
Purpose: Evaluate risk management effectiveness from risk profile assessment. Calculate risk management score based on mitigation coverage, test coverage for high risks, and risk score distribution.
Actions:
Load risk profile data (if available): total risks, critical/high/medium/low counts, mitigation status, test coverage
Calculate risk management score using formula:
Risk Management Score = (
(Mitigation Coverage × 0.5) +
(Test Coverage for High Risks × 0.3) +
(Risk Score Distribution × 0.2)
)
Determine risk management status (PASS ≥75%, CONCERNS 50-74%, FAIL <50%)
Identify risk management gaps (critical risks without mitigation, high risks without tests)
Calculate weighted contribution to overall gate (score × 25% weight)
See:references/templates.md#step-1-risk-management-dimension-output and gate-dimensions.md#risk-management
Step 2: Synthesize Test Coverage Dimension
Purpose: Evaluate test strategy and coverage from test-design assessment. Calculate test coverage score based on P0 tests passing, P1 tests implemented, and overall coverage percentage.
Actions:
Load test design data (if available): total tests by priority (P0/P1/P2), tests by level (unit/integration/e2e), implementation status, pass status, coverage percentage
Reference: See gate-dimensions.md for test coverage scoring methodology.
Step 3: Synthesize Traceability Dimension
Purpose: Evaluate requirements traceability from traceability assessment. Calculate traceability score based on implementation coverage, test coverage, and gap coverage.
Actions:
Load traceability data (if available): total AC, implementation status (implemented/partial/not implemented), test status (tested/partial/not tested), gaps by severity
Purpose: Evaluate non-functional requirements from NFR assessment. Use overall NFR score directly, check for critical gaps in security/reliability, and apply special rules for production blockers.
Actions:
Load NFR data (if available): overall score, category scores (security, performance, reliability, maintainability, scalability, usability), gaps by severity and category
NFR score is direct: Overall NFR Score from assessment (no additional calculation)
Determine NFR status (PASS ≥75%, CONCERNS 60-74%, FAIL <60%)
Purpose: Evaluate implementation quality from task specification and implementation record. Calculate implementation quality score based on task completion, testing, documentation, and code review status.
Actions:
Extract implementation data from task specification: status (Draft/In Progress/Review/Complete), files changed, implementation record (completion, testing, documentation)
Calculate implementation quality score using formula:
Purpose: Evaluate compliance with standards, policies, and regulatory requirements. Check if required compliance activities (security review, privacy review, etc.) have been completed.
Actions:
Check compliance requirements from config: standards (OWASP, WCAG, REST API best practices), policies (security review, privacy review, legal review)
Evaluate compliance: For each requirement, check if assessment addressed it and extract compliance status
Calculate compliance score: (Requirements Met / Total Requirements) × 100%
Determine compliance status (PASS 100%, CONCERNS 90-99%, FAIL <90%)
Calculate weighted contribution to overall gate (score × 5% weight)
Populate template from .claude/templates/quality-gate.md
Include all dimension details, scores, gaps, action items
Format for human readability with clear sections
Present summary to user:
Overall quality score and gate decision
Dimension scores with status indicators
Reasoning for decision
Critical issues and blockers (if any)
Action items with priorities
Waivers granted (if any)
Next steps (complete action items, proceed with merge, etc.)
Emit telemetry event with all metrics
Output: Assessment complete summary with overall score, dimension scores, gate decision, reasoning, reports generated (YAML/Markdown), next steps
See:references/templates.md#step-8-report-formats for complete YAML and Markdown report structures
See:gate-examples.md for example reports and summaries
1. Run risk-profile (optional)
2. Run test-design (optional)
3. Run trace-requirements (required*)
4. Run nfr-assess (required*)
5. Run quality-gate (synthesizes all)
*At least one of trace-requirements or nfr-assess required
