Files of Architecture optimization artifact templates
wondelai/
Show the full text158 lines
architecture-optimization artifact templates
Skeletons for every docs/ file this journey writes. Read the file before writing: if it is missing, create it from the full skeleton below (all section headings), then fill only the sections your phase names; if it exists, extend it — add or update your sections and preserve everyone else's. A journey that starts in a project with no docs/ folder creates several of these from scratch, which is the normal case for this skill.
Tracker: docs/ARCHITECTURE-OPTIMIZATION-PLAN.md
Created on the first run (Intake). Never shared with another journey.
# Architecture Optimization Plan
## Context
Intake answers, date started, hot paths in scope.
Mode: skills-installed | fallback (Briefs + references/methods.md) — set at the first phase, so a
resumed session knows whether earlier artifacts are skill-grade or Brief-grade.
## Phase Status
| Phase | Skill | Status | Artifact | Date |
|---|---|---|---|---|
| 1 | working-with-legacy-code | pending | TESTING.md, TECH-DEBT.md, PERFORMANCE.md | |
| 2 | clean-architecture | pending | ARCHITECTURE.md | |
| 3 | software-design-philosophy | pending | TECH-DEBT.md | |
| 4 | refactoring-patterns | pending | TECH-DEBT.md, TESTING.md | |
| 5 | system-design | pending | PERFORMANCE.md, ARCHITECTURE.md | |
| 6 | ddia-systems | pending | ARCHITECTURE.md, PERFORMANCE.md | |
| 7 | release-it | pending | RELIABILITY.md | |
| 8 | pragmatic-programmer | pending | PERFORMANCE.md, TECH-DEBT.md, TESTING.md | |
Statuses: pending · in-progress · awaiting-evidence · done · deferred: <reason> · skipped: <reason>
A phase parked at `awaiting-evidence` carries a Next Actions row naming the evidence owed, its owner,
and the date — otherwise the next session cannot tell what it is waiting for.
## Key Decisions
| Date | Phase | Decision | Rationale |
|---|---|---|---|
## Next Actions
- [ ] action (owner, due)
docs/PERFORMANCE.md
Measured performance state — baselines, budgets, findings, and the before/after ledger. Creates: architecture-optimization.
# Performance
## Baselines & Budgets
| Metric (p50/p95, throughput, cost) | Baseline (date) | Budget | Gate |
|---|---|---|---|
## Load Reality
Measured QPS average/peak, data volumes, growth rate. This is the single source for load numbers —
ARCHITECTURE.md `## System Context` cites it rather than repeating them.
## Profile Findings
| Hotspot | Evidence (profiler/APM) | Suspected cause | Fix | Status |
|---|---|---|---|---|
## Optimization Ledger
| Change | Before | After | Verdict (keep/revert) | Date |
|---|---|---|---|---|
docs/TESTING.md
The safety net — what behavior is pinned, where the gaps are.
# Testing
## Test Strategy
Pyramid, tooling, what "green" gates.
## Safety Net Map
| Module | Pinned behaviors | Test files | Gaps |
|---|---|---|---|
## Characterization Backlog
- [ ] module (risk, priority)
## CI Gates
docs/TECH-DEBT.md
The debt ledger — single queue for all code journeys.
# Technical Debt
## Debt Ledger
| Item | Location | Type | Risk | Effort | Priority | Status |
|---|---|---|---|---|---|---|
## Smell Inventory
| Smell | Location | Refactoring | Status |
|---|---|---|---|
## Sprout / Wrap Register
Code added beside legacy (to fold back in later).
## Debt Budget & Broken-Windows Policy
Time per iteration; what gets fixed now vs boarded up with a ticket.
## Adopted Conventions
docs/ARCHITECTURE.md
System structure and decisions — includes domain model and data decisions (no separate DATA.md or DOMAIN.md).
# Architecture
## System Context
What the system does, key integrations, load reality (cite PERFORMANCE.md `## Load Reality`).
## Layer Map & Dependency Rule
Layers, what depends on what, current violations.
| Violation | Location | Fix | Status |
|---|---|---|---|
## Bounded Contexts & Context Map
Contexts, relationships, anti-corruption layers.
## Domain Glossary (Ubiquitous Language)
| Term | Meaning | Code name |
|---|---|---|
## Data & Storage Decisions
Data models, storage engines, isolation levels, replication, system-of-record vs derived data.
## Decision Log
| Date | Decision | Why | Alternatives rejected |
|---|---|---|---|
docs/RELIABILITY.md
Production hardening status.
# Reliability
## Integration-Point Audit
| Dependency | Timeout | Circuit breaker | Bulkhead | Retry policy | Status |
|---|---|---|---|---|---|
## Query & Resource Findings
Unbounded result sets, missing LIMITs/pagination, blocked threads.
## Health Checks & Metrics
Deep health checks · RED metrics · symptom-based alerts.
## Deploy vs Release
Feature flags, expand-contract migrations, rollback plan.
| 1 | # architecture-optimization artifact templates |
| 2 | |
| 3 | Skeletons for every docs/ file this journey writes. Read the file before writing: if it is missing, create it from the full skeleton below (all section headings), then fill only the sections your phase names; if it exists, extend it — add or update your sections and preserve everyone else's. A journey that starts in a project with no `docs/` folder creates several of these from scratch, which is the normal case for this skill. |
| 4 | |
| 5 | ## Tracker: docs/ARCHITECTURE-OPTIMIZATION-PLAN.md |
| 6 | |
| 7 | Created on the first run (Intake). Never shared with another journey. |
| 8 | |
| 9 | |
| 10 | # Architecture Optimization Plan |
| 11 | |
| 12 | ## Context |
| 13 | Intake answers, date started, hot paths in scope. |
| 14 | Mode: skills-installed | fallback (Briefs + references/methods.md) — set at the first phase, so a |
| 15 | resumed session knows whether earlier artifacts are skill-grade or Brief-grade. |
| 16 | |
| 17 | ## Phase Status |
| 18 | | Phase | Skill | Status | Artifact | Date | |
| 19 | |---|---|---|---|---| |
| 20 | | 1 | working-with-legacy-code | pending | TESTING.md, TECH-DEBT.md, PERFORMANCE.md | | |
| 21 | | 2 | clean-architecture | pending | ARCHITECTURE.md | | |
| 22 | | 3 | software-design-philosophy | pending | TECH-DEBT.md | | |
| 23 | | 4 | refactoring-patterns | pending | TECH-DEBT.md, TESTING.md | | |
| 24 | | 5 | system-design | pending | PERFORMANCE.md, ARCHITECTURE.md | | |
| 25 | | 6 | ddia-systems | pending | ARCHITECTURE.md, PERFORMANCE.md | | |
| 26 | | 7 | release-it | pending | RELIABILITY.md | | |
| 27 | | 8 | pragmatic-programmer | pending | PERFORMANCE.md, TECH-DEBT.md, TESTING.md | | |
| 28 | Statuses: pending · in-progress · awaiting-evidence · done · deferred: <reason> · skipped: <reason> |
| 29 | A phase parked at `awaiting-evidence` carries a Next Actions row naming the evidence owed, its owner, |
| 30 | and the date — otherwise the next session cannot tell what it is waiting for. |
| 31 | |
| 32 | ## Key Decisions |
| 33 | | Date | Phase | Decision | Rationale | |
| 34 | |---|---|---|---| |
| 35 | |
| 36 | ## Next Actions |
| 37 | - [ ] action (owner, due) |
| 38 | |
| 39 | |
| 40 | ## docs/PERFORMANCE.md |
| 41 | |
| 42 | Measured performance state — baselines, budgets, findings, and the before/after ledger. Creates: architecture-optimization. |
| 43 | |
| 44 | |
| 45 | # Performance |
| 46 | |
| 47 | ## Baselines & Budgets |
| 48 | | Metric (p50/p95, throughput, cost) | Baseline (date) | Budget | Gate | |
| 49 | |---|---|---|---| |
| 50 | |
| 51 | ## Load Reality |
| 52 | Measured QPS average/peak, data volumes, growth rate. This is the single source for load numbers — |
| 53 | ARCHITECTURE.md `## System Context` cites it rather than repeating them. |
| 54 | |
| 55 | ## Profile Findings |
| 56 | | Hotspot | Evidence (profiler/APM) | Suspected cause | Fix | Status | |
| 57 | |---|---|---|---|---| |
| 58 | |
| 59 | ## Optimization Ledger |
| 60 | | Change | Before | After | Verdict (keep/revert) | Date | |
| 61 | |---|---|---|---|---| |
| 62 | |
| 63 | |
| 64 | ## docs/TESTING.md |
| 65 | |
| 66 | The safety net — what behavior is pinned, where the gaps are. |
| 67 | |
| 68 | |
| 69 | # Testing |
| 70 | |
| 71 | ## Test Strategy |
| 72 | Pyramid, tooling, what "green" gates. |
| 73 | |
| 74 | ## Safety Net Map |
| 75 | | Module | Pinned behaviors | Test files | Gaps | |
| 76 | |---|---|---|---| |
| 77 | |
| 78 | ## Characterization Backlog |
| 79 | - [ ] module (risk, priority) |
| 80 | |
| 81 | ## CI Gates |
| 82 | |
| 83 | |
| 84 | ## docs/TECH-DEBT.md |
| 85 | |
| 86 | The debt ledger — single queue for all code journeys. |
| 87 | |
| 88 | |
| 89 | # Technical Debt |
| 90 | |
| 91 | ## Debt Ledger |
| 92 | | Item | Location | Type | Risk | Effort | Priority | Status | |
| 93 | |---|---|---|---|---|---|---| |
| 94 | |
| 95 | ## Smell Inventory |
| 96 | | Smell | Location | Refactoring | Status | |
| 97 | |---|---|---|---| |
| 98 | |
| 99 | ## Sprout / Wrap Register |
| 100 | Code added beside legacy (to fold back in later). |
| 101 | |
| 102 | ## Debt Budget & Broken-Windows Policy |
| 103 | Time per iteration; what gets fixed now vs boarded up with a ticket. |
| 104 | |
| 105 | ## Adopted Conventions |
| 106 | |
| 107 | |
| 108 | ## docs/ARCHITECTURE.md |
| 109 | |
| 110 | System structure and decisions — includes domain model and data decisions (no separate DATA.md or DOMAIN.md). |
| 111 | |
| 112 | |
| 113 | # Architecture |
| 114 | |
| 115 | ## System Context |
| 116 | What the system does, key integrations, load reality (cite PERFORMANCE.md `## Load Reality`). |
| 117 | |
| 118 | ## Layer Map & Dependency Rule |
| 119 | Layers, what depends on what, current violations. |
| 120 | | Violation | Location | Fix | Status | |
| 121 | |---|---|---|---| |
| 122 | |
| 123 | ## Bounded Contexts & Context Map |
| 124 | Contexts, relationships, anti-corruption layers. |
| 125 | |
| 126 | ## Domain Glossary (Ubiquitous Language) |
| 127 | | Term | Meaning | Code name | |
| 128 | |---|---|---| |
| 129 | |
| 130 | ## Data & Storage Decisions |
| 131 | Data models, storage engines, isolation levels, replication, system-of-record vs derived data. |
| 132 | |
| 133 | ## Decision Log |
| 134 | | Date | Decision | Why | Alternatives rejected | |
| 135 | |---|---|---|---| |
| 136 | |
| 137 | |
| 138 | ## docs/RELIABILITY.md |
| 139 | |
| 140 | Production hardening status. |
| 141 | |
| 142 | |
| 143 | # Reliability |
| 144 | |
| 145 | ## Integration-Point Audit |
| 146 | | Dependency | Timeout | Circuit breaker | Bulkhead | Retry policy | Status | |
| 147 | |---|---|---|---|---|---| |
| 148 | |
| 149 | ## Query & Resource Findings |
| 150 | Unbounded result sets, missing LIMITs/pagination, blocked threads. |
| 151 | |
| 152 | ## Health Checks & Metrics |
| 153 | Deep health checks · RED metrics · symptom-based alerts. |
| 154 | |
| 155 | ## Deploy vs Release |
| 156 | Feature flags, expand-contract migrations, rollback plan. |
| 157 | |
| 158 |
Discussion
Alternatives
Browse more free Claude skills or everything in Development.