gh-yaleh-meta-cc-claude/skills/methodology-bootstrapping/examples/iteration-structure-template.md

Iteration N: [Iteration Title]

Date: YYYY-MM-DD Duration: ~X hours Status: [In Progress / Completed] Framework: BAIME (Bootstrapped AI Methodology Engineering)

---

1. Executive Summary

[2-3 paragraphs summarizing:]

• Iteration focus and primary objectives
• Key achievements and deliverables
• Key learnings and insights
• Value scores with gaps to target

Value Scores:

• V_instance(s_N) = [X.XX] (Target: 0.80, Gap: [±X.XX])
• V_meta(s_N) = [X.XX] (Target: 0.80, Gap: [±X.XX])

---

2. Pre-Execution Context

Previous State (s_{N-1}): From Iteration N-1

• V_instance(s_{N-1}) = [X.XX] (Target: 0.80, Gap: [±X.XX])
  • [Component 1] = [X.XX]
  • [Component 2] = [X.XX]
  • [Component 3] = [X.XX]
  • [Component 4] = [X.XX]
• V_meta(s_{N-1}) = [X.XX] (Target: 0.80, Gap: [±X.XX])
  • V_completeness = [X.XX]
  • V_effectiveness = [X.XX]
  • V_reusability = [X.XX]

Meta-Agent: M_{N-1} ([describe stability status, e.g., "M₀ stable, 5 capabilities"])

Agent Set: A_{N-1} = {[list agent names]} ([describe type, e.g., "generic agents" or "2 specialized"])

Primary Objectives:

1. [Objective 1 with success indicator: ✅/⚠️/❌]
2. [Objective 2 with success indicator: ✅/⚠️/❌]
3. [Objective 3 with success indicator: ✅/⚠️/❌]
4. [Objective 4 with success indicator: ✅/⚠️/❌]

---

3. Work Executed

Phase 1: OBSERVE - [Description] (~X min/hours)

Data Collection:

Analysis:

• [Finding 1 Title]: [Detailed finding with data]
• [Finding 2 Title]: [Detailed finding with data]
• [Finding 3 Title]: [Detailed finding with data]

Gaps Identified:

• [Gap area 1]: [Current state] → [Target state]
• [Gap area 2]: [Current state] → [Target state]
• [Gap area 3]: [Current state] → [Target state]

Phase 2: CODIFY - [Description] (~X min/hours)

Deliverable: [path/to/knowledge-file.md] ([X lines])

Content Structure:

Patterns Extracted:

• [Pattern 1 Name]: [Description, applicability, benefits]
• [Pattern 2 Name]: [Description, applicability, benefits]

Decision Made: [Key decision with rationale]

Rationale:

• [Reason 1]
• [Reason 2]
• [Reason 3]

Phase 3: AUTOMATE - [Description] (~X min/hours)

Approach: [High-level approach description]

Changes Made:

1. [Change Category 1]:

   • [Specific change 1a]
   • [Specific change 1b]

2. [Change Category 2]:

   • [Specific change 2a]
   • [Specific change 2b]

3. [Change Category 3]:

   // Example code changes
   // Before:
   [old code]
   
   // After:
   [new code]
   

Code Changes:

• Modified: [file path] ([X lines changed], [description])
• Created: [file path] ([X lines], [description])

Results:

Before: [metric]
After:  [metric]

Benefits:

• ✅ [Benefit 1 with evidence]
• ✅ [Benefit 2 with evidence]
• ✅ [Benefit 3 with evidence]

Phase 4: EVALUATE - Calculate V(s_N) (~X min/hours)

Measurements:

• [Metric 1]: [baseline value] → [final value] (change: [±X%])
• [Metric 2]: [baseline value] → [final value] (change: [±X%])
• [Metric 3]: [baseline value] → [final value] (change: [±X%])

Why [Metric Changed/Didn't Change]:

• [Reason 1]
• [Reason 2]

---

4. Value Calculations

V_instance(s_N) Calculation

Formula:

V_instance(s) = [weight1]·[Component1] + [weight2]·[Component2] + [weight3]·[Component3] + [weight4]·[Component4]

Component 1: [Component Name]

Measurement:

Score: [X.XX] ([±X.XX from previous iteration])

Evidence:

• [Concrete evidence 1 with data]
• [Concrete evidence 2 with data]
• [Concrete evidence 3 with data]

Component 2: [Component Name]

Measurement:

Score: [X.XX] ([±X.XX from previous iteration])

Evidence:

• [Concrete evidence 1]
• [Concrete evidence 2]

Component 3: [Component Name]

Measurement:

Score: [X.XX] ([±X.XX from previous iteration])

Evidence:

• [Concrete evidence 1]

Component 4: [Component Name]

Measurement: [Description]

Score: [X.XX] ([±X.XX from previous iteration])

Evidence: [Concrete evidence]

V_instance(s_N) Final Calculation

V_instance(s_N) = [weight1]·([score1]) + [weight2]·([score2]) + [weight3]·([score3]) + [weight4]·([score4])
               = [term1] + [term2] + [term3] + [term4]
               = [sum]
               ≈ [X.XX]

V_instance(s_N) = [X.XX] (Target: 0.80, Gap: [±X.XX] or [±X]%)

Change from s_{N-1}: [±X.XX] ([±X]% improvement/decline)

---

V_meta(s_N) Calculation

Formula:

V_meta(s) = 0.40·V_completeness + 0.30·V_effectiveness + 0.30·V_reusability

Component 1: V_completeness (Methodology Documentation)

Checklist Progress ([X]/15 items):

• Process steps documented ✅
• Decision criteria defined ✅
• Examples provided ✅
• Edge cases covered ✅
• Failure modes documented ✅
• Rationale explained ✅
• [Additional item 7]
• [Additional item 8]
• [Additional item 9]
• [Additional item 10]
• [Additional item 11]
• [Additional item 12]
• [Additional item 13]
• [Additional item 14]
• [Additional item 15]

Score: [X.XX] ([±X.XX from previous iteration])

Evidence:

• [Evidence 1: document created, X lines]
• [Evidence 2: patterns added]
• [Evidence 3: examples provided]

Gap to 1.0: Still missing [X]/15 items

• [Missing item 1]
• [Missing item 2]
• [Missing item 3]

Component 2: V_effectiveness (Practical Impact)

Measurement:

• Time savings: [X hours for task] (vs [Y hours ad-hoc] → [Z]x speedup)
• Pattern usage: [Describe how patterns were applied]
• Quality improvement: [Metric] improved from [X] to [Y]
• Speedup estimate: [Z]x faster than ad-hoc approach

Score: [X.XX] ([±X.XX from previous iteration])

Evidence:

• [Evidence 1: time measurement]
• [Evidence 2: quality improvement]
• [Evidence 3: pattern effectiveness]

Gap to 0.80: [What's needed]

• [Gap item 1]
• [Gap item 2]

Component 3: V_reusability (Transferability)

Assessment: [Overall transferability assessment]

Score: [X.XX] ([±X.XX from previous iteration])

Evidence:

• [Evidence 1: universal patterns identified]
• [Evidence 2: language-agnostic concepts]
• [Evidence 3: cross-domain applicability]

Transferability Estimate:

• Same language ([language]): ~[X]% modification ([reason])
• Similar language ([language] → [language]): ~[X]% modification ([reason])
• Different paradigm ([language] → [language]): ~[X]% modification ([reason])

Gap to 0.80: [What's needed]

• [Gap item 1]
• [Gap item 2]

V_meta(s_N) Final Calculation

V_meta(s_N) = 0.40·([completeness]) + 0.30·([effectiveness]) + 0.30·([reusability])
           = [term1] + [term2] + [term3]
           = [sum]
           ≈ [X.XX]

V_meta(s_N) = [X.XX] (Target: 0.80, Gap: [±X.XX] or [±X]%)

Change from s_{N-1}: [±X.XX] ([±X]% improvement/decline)

---

5. Gap Analysis

Instance Layer Gaps (ΔV = [±X.XX] to target)

Status: [Assessment, e.g., "🔄 MODERATE PROGRESS (X% of target)"]

Priority 1: [Gap Area] ([Component] = [X.XX], need [±X.XX])

• [Action item 1]: [Details, expected impact]
• [Action item 2]: [Details, expected impact]
• [Action item 3]: [Details, expected impact]

Priority 2: [Gap Area] ([Component] = [X.XX], need [±X.XX])

• [Action item 1]
• [Action item 2]

Priority 3: [Gap Area] ([Component] = [X.XX], status)

• [Action item 1]

Priority 4: [Gap Area] ([Component] = [X.XX], status)

• [Assessment]

Estimated Work: [X] more iteration(s) to reach V_instance ≥ 0.80

Meta Layer Gaps (ΔV = [±X.XX] to target)

Status: [Assessment]

Priority 1: Completeness (V_completeness = [X.XX], need [±X.XX])

• [Action item 1]
• [Action item 2]
• [Action item 3]

Priority 2: Effectiveness (V_effectiveness = [X.XX], need [±X.XX])

• [Action item 1]
• [Action item 2]
• [Action item 3]

Priority 3: Reusability (V_reusability = [X.XX], need [±X.XX])

• [Action item 1]
• [Action item 2]
• [Action item 3]

Estimated Work: [X] more iteration(s) to reach V_meta ≥ 0.80

---

6. Convergence Check

Criteria Assessment

Dual Threshold:

• V_instance(s_N) ≥ 0.80: [✅ YES / ❌ NO] ([X.XX], gap: [±X.XX], [X]% of target)
• V_meta(s_N) ≥ 0.80: [✅ YES / ❌ NO] ([X.XX], gap: [±X.XX], [X]% of target)

System Stability:

• M_N == M_{N-1}: [✅ YES / ❌ NO] ([rationale, e.g., "M₀ stable, no evolution needed"])
• A_N == A_{N-1}: [✅ YES / ❌ NO] ([rationale, e.g., "generic agents sufficient"])

Objectives Complete:

• [Objective 1]: [✅ YES / ❌ NO] ([status])
• [Objective 2]: [✅ YES / ❌ NO] ([status])
• [Objective 3]: [✅ YES / ❌ NO] ([status])
• [Objective 4]: [✅ YES / ❌ NO] ([status])

Diminishing Returns:

• ΔV_instance = [±X.XX] ([assessment, e.g., "small but positive", "diminishing"])
• ΔV_meta = [±X.XX] ([assessment])
• [Overall assessment]

Status: [✅ CONVERGED / ❌ NOT CONVERGED]

Reason:

• [Detailed rationale for convergence decision]
• [Supporting evidence 1]
• [Supporting evidence 2]

Progress Trajectory:

• Instance layer: [s0] → [s1] → [s2] → ... → [sN]
• Meta layer: [s0] → [s1] → [s2] → ... → [sN]

Estimated Iterations to Convergence: [X] more iteration(s)

• Iteration N+1: [Expected progress]
• Iteration N+2: [Expected progress]
• Iteration N+3: [Expected progress]

---

7. Evolution Decisions

Agent Evolution

Current Agent Set: A_N = [list agents, e.g., "A_{N-1}" if unchanged]

Sufficiency Analysis:

• [✅/❌] [Agent 1 name]: [Performance assessment]
• [✅/❌] [Agent 2 name]: [Performance assessment]
• [✅/❌] [Agent 3 name]: [Performance assessment]

Decision: [✅ NO EVOLUTION NEEDED / ⚠️ EVOLUTION NEEDED]

Rationale:

• [Reason 1]
• [Reason 2]
• [Reason 3]

If Evolution: [Describe new agent, rationale, expected improvement]

Re-evaluate: [When to reassess, e.g., "After Iteration N+1 if [condition]"]

Meta-Agent Evolution

Current Meta-Agent: M_N = [describe, e.g., "M_{N-1} (5 capabilities)"]

Sufficiency Analysis:

• [✅/❌] [Capability 1]: [Effectiveness assessment]
• [✅/❌] [Capability 2]: [Effectiveness assessment]
• [✅/❌] [Capability 3]: [Effectiveness assessment]
• [✅/❌] [Capability 4]: [Effectiveness assessment]
• [✅/❌] [Capability 5]: [Effectiveness assessment]

Decision: [✅ NO EVOLUTION NEEDED / ⚠️ EVOLUTION NEEDED]

Rationale: [Detailed reasoning]

If Evolution: [Describe new capability, rationale, expected improvement]

---

8. Artifacts Created

Data Files

• [path/to/data-file-1] - [Description, e.g., "Test coverage report (X%)"]
• [path/to/data-file-2] - [Description]
• [path/to/data-file-3] - [Description]

Knowledge Files

• [path/to/knowledge-file-1] - [Description, e.g., "X lines, Pattern Y documented"]
• [path/to/knowledge-file-2] - [Description]

Code Changes

• Modified: [file path] ([X lines, description])
• Created: [file path] ([X lines, description])
• Deleted: [file path] ([reason])

Other Artifacts

---

9. Reflections

What Worked

1. [Success 1 Title]: [Detailed description with evidence]
2. [Success 2 Title]: [Detailed description with evidence]
3. [Success 3 Title]: [Detailed description with evidence]
4. [Success 4 Title]: [Detailed description with evidence]

What Didn't Work

1. [Challenge 1 Title]: [Detailed description with root cause]
2. [Challenge 2 Title]: [Detailed description with root cause]
3. [Challenge 3 Title]: [Detailed description with root cause]

Learnings

1. [Learning 1 Title]: [Insight gained, applicability]
2. [Learning 2 Title]: [Insight gained, applicability]
3. [Learning 3 Title]: [Insight gained, applicability]
4. [Learning 4 Title]: [Insight gained, applicability]

Insights for Methodology

1. [Insight 1 Title]: [Meta-level insight for methodology development]
2. [Insight 2 Title]: [Meta-level insight for methodology development]
3. [Insight 3 Title]: [Meta-level insight for methodology development]
4. [Insight 4 Title]: [Meta-level insight for methodology development]

---

10. Conclusion

[Comprehensive summary paragraph covering:]

• Overall iteration assessment
• Key metrics and their changes
• Critical decisions made and their rationale
• Methodology development progress

Key Metrics:

• [Metric 1]: [value] ([change], target: [target])
• [Metric 2]: [value] ([change], target: [target])
• [Metric 3]: [value] ([change], target: [target])

Value Functions:

• V_instance(s_N) = [X.XX] ([X]% of target, [±X.XX] improvement)
• V_meta(s_N) = [X.XX] ([X]% of target, [±X.XX] improvement - [±X]% growth)

Key Insight: [Main takeaway from this iteration in 1-2 sentences]

Critical Decision: [Most important decision made and its impact]

Next Steps: [What Iteration N+1 will focus on, expected outcomes]

Confidence: [Assessment of confidence in achieving next iteration goals, e.g., "High / Medium / Low" with reasoning]

---

Status: [Status indicator, e.g., "✅ [Achievement]" or "🔄 [In Progress]"] Next: Iteration N+1 - [Focus Area] Expected Duration: [X] hours