Execute-Quality 150 Protocol

Core Principle: Commit to excellence. Work at 150% — 100% core delivery + 50% strengthening through verification, alternatives, and risk awareness.

What This Skill Does

When you invoke this skill, you're asking AI to:

• Work deeper — Full context, not fragments
• Think harder — Verify assumptions, consider alternatives
• Validate more — Check at every step, not just at the end
• Deliver better — With confidence levels and documented reasoning

The 150% Quality Standard

Dimension	100% Core	+50% Strengthening
Context	Understand the task	+ Understand dependencies and boundaries
Research	Find answers	+ Verify from multiple sources
Assumptions	Use them	+ Test and validate them
Solutions	Provide one	+ Consider alternatives
Risks	Ignore	+ Identify and mitigate
Confidence	Implicit	+ Explicit with percentages

Execution Protocol

Step 1: ASSESS

Evaluate what maximum quality means for THIS task:

• What's the complexity level?
• What context needs full reading?
• What assumptions need checking?
• What could go wrong?

Step 2: COMMIT

Declare your quality approach:

🎯 **Max-Quality 150 Engaged**

For this task, maximum quality means:
- [What I'll read fully]
- [What I'll verify]
- [What alternatives I'll consider]
- [What risks I'll check]

Step 3: EXECUTE

Work with 150% coverage:

• Read fully — Complete files, not snippets
• Verify assumptions — Don't trust, confirm
• Consider alternatives — At least 3 approaches
• Check risks — What could fail?
• Validate continuously — Not just at the end

Step 4: DELIVER

Provide result with quality indicators:

• Confidence level (percentage)
• What was verified vs assumed
• Known limitations
• Recommendations

Quality Behaviors

When max-quality-150 is active:

Reading & Research

❌ Standard: Read first 50 lines of file
✅ Max-Quality: Read entire file + imports + dependencies

Assumptions

❌ Standard: "This function probably handles X"
✅ Max-Quality: "Verified: function handles X (line 45-67)"

Solutions

❌ Standard: "Here's the fix"
✅ Max-Quality: "Considered 3 approaches, recommending A because..."

Validation

❌ Standard: Check if it compiles
✅ Max-Quality: Test edge cases, verify integration, check performance

Output Format

When delivering with max-quality-150:

🎯 **Max-Quality 150 Delivery**

**Task:** [What was requested]

**Quality Approach Used:**
- ✅ Full context read: [files/systems]
- ✅ Assumptions verified: [list]
- ✅ Alternatives considered: [count]
- ✅ Risks assessed: [count]

**Result:**
[Detailed, high-quality output]

**Confidence:** [X]%
**Verified:** [What was confirmed]
**Assumed:** [What remains unverified]
**Limitations:** [Known gaps]

When to Use This Skill

Invoke max-quality-150 for:

• Critical tasks — Production deployments, security changes
• Complex problems — Multi-system, many dependencies
• Important decisions — Architecture, strategy, major choices
• When stakes are high — User data, business logic, integrations
• When you need confidence — Can't afford to be wrong

Operational Rules

1. FULL CONTEXT: Read complete files, not fragments
2. VERIFY ASSUMPTIONS: Don't trust, confirm with evidence
3. MULTIPLE PERSPECTIVES: Consider at least 3 approaches
4. CONTINUOUS VALIDATION: Check quality at every step
5. EXPLICIT CONFIDENCE: State certainty levels numerically
6. DOCUMENT REASONING: Explain why, not just what
7. ACKNOWLEDGE LIMITS: State what you don't know

Examples

❌ Standard Quality

User: "Review this authentication code"
AI: "The code looks fine. Consider adding input validation."

✅ Max-Quality 150

User: "Review this authentication code"
AI: "🎯 Max-Quality 150 Engaged

For this security-critical review, I'll:
- Read full auth module + dependencies
- Check against OWASP standards
- Verify session handling
- Test edge cases mentally

[After thorough review]

🎯 Max-Quality 150 Delivery

Confidence: 92%

Verified Issues:
1. Password not hashed (line 45) - Critical
2. No rate limiting (lines 78-92) - High
3. Session token predictable (line 112) - Critical

Verified Safe:
- Input sanitization present (line 23)
- HTTPS enforced (config verified)

Alternatives Considered:
- bcrypt vs argon2 for hashing
- Token-based vs session-based auth

Risks Identified:
- Current code vulnerable to brute force
- Session hijacking possible

Recommendations: [detailed fixes]"

Failure Modes & Recovery

Failure	Detection	Recovery
Shallow work	Skipping full context	Stop, read completely
Unverified claims	Assumptions without evidence	Go verify before proceeding
Single solution	No alternatives considered	Generate 3+ options
Missing risks	No failure analysis	Do risk assessment
Low confidence	<85% certainty	More research or flag to user

Relationship to Other Skills

• deep-think-150 — For quality reasoning (thinking)
• max-quality-150 — For quality execution (doing)

Use both when task requires excellent thinking AND excellent execution.

Session Log Entry (MANDATORY)

After completing this skill, write to

.sessions/SESSION_[date]-[name].md

:

### [HH:MM] Execute-Quality 150 Complete
**Task:** <task executed>
**Quality Score:** <confidence %>
**Verified:** <what was verified>
**Artifacts:** <files delivered>

---

Remember: max-quality-150 is not about being slow — it's about being thorough. The extra effort upfront prevents costly mistakes later. Quality is an investment, not an expense.