diegosouzapw Validate-Integrity 150 Protocol
Core Principle: Check yourself before delivery. Verify goal alignment, completeness, correctness. Identify what could still go wrong. Score the quality honestly.
What This Skill Does
When you invoke this skill, you're asking AI to:
- Verify alignment — Does result match original goal?
- Check completeness — Are all requirements satisfied?
- Validate correctness — Is it technically accurate?
- Assess risks — What residual issues remain?
- Score quality — Honest 0-100 rating with reasoning
The 150% Check Rule
| Dimension | 100% Core | +50% Enhancement |
|---|---|---|
| Goal | Meets objectives | + Exceeds expectations |
| Complete | Requirements done | + Edge cases covered |
| Correct | Technically valid | + Best practices followed |
| Risks | Known issues noted | + Mitigation strategies ready |
Quality Assessment Framework
CORE INTEGRITY (100% Required)
├── Goal Alignment: Meets original objectives
├── Completeness: All requirements satisfied
├── Correctness: Technically accurate and valid
└── Consistency: Internal logic coherent
EXTENDED VALIDATION (50% Enhancement)
├── Context Fit: Works in broader environment
├── Side Effects: No unintended consequences
├── Edge Cases: Handles unusual scenarios
└── Future-Proofing: Adaptable to changes
Quality Score Framework
| Score | Rating | Description | Action |
|---|---|---|---|
| 95-100 | Excellent | Exceeds expectations | Deliver with confidence |
| 85-94 | Good | Meets all requirements | Deliver, note improvements |
| 75-84 | Acceptable | Minor issues exist | Deliver with caveats |
| 60-74 | Needs Work | Significant issues | Fix before delivery |
| <60 | Unacceptable | Major problems | Return to development |
When to Use This Skill
- Before delivery — Any significant work output
- After implementation — Code, features, documents
- Before handoff — To another person or system
- When uncertain — Need confidence check
- After major changes — Verify nothing broke
Execution Protocol
Step 1: GOAL ALIGNMENT
🎯 **GOAL VERIFICATION**
**Original Objective:** [What was requested]
**Delivered Result:** [What was produced]
**Alignment:** ✅ Match | ⚠️ Partial | ❌ Mismatch
**Gaps:** [Any differences from original goal]
Step 2: COMPLETENESS ASSESSMENT
📋 **COMPLETENESS CHECK**
Requirements:
- [ ] Requirement 1: [Status]
- [ ] Requirement 2: [Status]
- [ ] Requirement 3: [Status]
**Coverage:** [X]% of requirements met
**Missing:** [What's not done]
Step 3: CORRECTNESS VALIDATION
✓ **CORRECTNESS CHECK**
**Technical Accuracy:** [Assessment]
**Logic Coherence:** [Assessment]
**Best Practices:** [Assessment]
**Issues Found:** [List any problems]
Step 4: CONTEXT INTEGRATION
🔗 **CONTEXT FIT**
**Environment:** Works in target environment?
**Integration:** Properly connected to dependencies?
**Side Effects:** Any unintended consequences?
**Performance:** Meets performance requirements?
Step 5: RISK EVALUATION
⚠️ **RISK ASSESSMENT**
**Known Issues:**
- [Issue 1]: [Severity] - [Mitigation]
- [Issue 2]: [Severity] - [Mitigation]
**Edge Cases:** [Unusual scenarios covered?]
**Security:** [Any vulnerabilities?]
**Scalability:** [Can handle expected load?]
Step 6: SELF-CRITIQUE
🔮 **SELF-CRITIQUE**
**What could be better?**
- [Improvement 1]
- [Improvement 2]
**Alternative approaches considered?**
- [Alternative and why not chosen]
**Lessons for future:**
- [Learning 1]
Step 7: FINAL SCORE
Calculate and declare:
📊 **QUALITY SCORE:** [X]/100 ([Rating])
**Reasoning:** [Why this score]
Output Format
🔍 **INTEGRITY-CHECK 150 COMPLETE**
**Quality Score:** [X]/100 ([Rating])
**✅ STRENGTHS:**
- [Key positive finding 1]
- [Key positive finding 2]
- [Quality achievement]
**⚠️ AREAS FOR ATTENTION:**
- [Minor issue 1]
- [Recommendation 1]
- [Residual risk 1]
**🚫 CRITICAL ISSUES:** [None / List blockers]
**📋 DELIVERY STATUS:**
[✅ Ready | ⚠️ Conditional | ❌ Not Ready]
**🎯 CONFIDENCE LEVEL:** [High/Medium/Low]
**Reasoning:** [Why this confidence]
**RECOMMENDATIONS:**
- [What to do next]
- [What to monitor]
Integrity Checklist
🎯 CORE QUALITY
- Goal Achievement: Original objectives fully met?
- Requirement Coverage: All specified requirements?
- Technical Accuracy: Solution correct?
- Consistency: Internal logic coherent?
🔍 CONTEXT
- Environment Fit: Works in target environment?
- Integration: Properly connected?
- Side Effects: None unintended?
- Performance: Meets requirements?
⚠️ RISKS
- Known Issues: Documented and acceptable?
- Edge Cases: Unusual scenarios handled?
- Security: No vulnerabilities?
- Scalability: Handles expected load?
🔮 FUTURE
- Maintainability: Easy to maintain?
- Documentation: Well-documented?
- Knowledge Transfer: Others can understand?
Operational Rules
- MANDATORY CHECK: Every significant work needs integrity check
- OBJECTIVE CRITIQUE: Self-assess without bias
- COMPLETE COVERAGE: Check all quality dimensions
- RISK TRANSPARENCY: Document residual risks honestly
- IMPROVEMENT FOCUS: Identify what could be better
- HONEST SCORING: Don't inflate quality score
Failure Modes & Recovery
| Failure | Detection | Recovery |
|---|---|---|
| Overconfidence | Score too high | Get external review |
| Incomplete Check | Missed aspects | Use full checklist |
| Bias Blindness | Ignoring own errors | Delayed assessment |
| Risk Denial | Ignoring issues | Explicit risk documentation |
Examples
❌ Without Integrity-Check
AI: "Done! Feature implemented."
[Deploys to production]
Result: Critical security vulnerability discovered by users
✅ With Integrity-Check 150
🔍 INTEGRITY-CHECK 150 COMPLETE
Quality Score: 88/100 (Good)
✅ STRENGTHS:
- User registration fully functional
- Mobile/desktop compatibility verified
- All 5 requirements implemented
- Unit tests passing (94% coverage)
⚠️ AREAS FOR ATTENTION:
- Loading states could be smoother
- Error messages need localization
- Consider rate limiting for production
🚫 CRITICAL ISSUES: None
📋 DELIVERY STATUS: ✅ Ready
🎯 CONFIDENCE: High
Reasoning: All core requirements met, tested on multiple
devices, no blocking issues found.
RECOMMENDATIONS:
- Add loading animation polish in next sprint
- Set up error message translation pipeline
- Implement rate limiting before high traffic events
Relationship to Other Skills
- gated-exec-150 → Executes the plan
- integrity-check-150 → Validates the result
- 74-mid-session-save-150 → Documents continuity via checkpoints
Session Log Entry (MANDATORY)
After completing this skill, write to .sessions/SESSION_[date]-[name].md:
### [HH:MM] Validate-Integrity 150 Complete
**Target:** <what was validated>
**Score:** <X/100>
**Issues:** <found issues/none>
**Recommendation:** <Deliver/Fix>
Remember: Integrity-check isn't criticism — it's quality assurance. An honest 85 score with documented issues is more valuable than a false 100. The check protects you and the user from preventable problems.