メニュー
Maps project constraints (budget, team size, timeline, ops maturity, compliance) to architecture decisions
デフォルトでは、最初にソースを確認する Prompt が選択されています。直接コマンドに切り替えるか、ローカルコピーをダウンロードすることもできます。
インストールを決める前に、SKILL.md と SkillsMP に表示されている付属ファイルをお読みください。
Codex または Claude でインストール この Prompt をコピーして Codex、Claude、または他のアシスタントに貼り付けると、Skill ページを確認してインストールできます。
直接コマンドでは確認用 Prompt が省略されます。実行前にソースを確認してください。
npx skills add https://github.com/navraj007in/architecture-cowork-plugin --skill constraint-solverコマンドは1行のまま表示されます。コピー前に横へスクロールして全体を確認してください。
ローカルで確認しますか？SkillsMP が現在取得できるファイルをダウンロードできます。
Identifies missing prerequisites (blockers) for each /architect: command based on _state.json field requirements
Tracks actual infrastructure costs against estimated costs and alerts when spend drifts from projection
Defines when and how commands must log architecture decisions to _state.json decisions[] for auditability
SKILL.md を表示中
SOC 職業分類に基づく
| name | constraint-solver |
| description | Maps project constraints (budget, team size, timeline, ops maturity, compliance) to architecture decisions |
Maps project constraints (budget, team size, timeline, ops maturity, compliance) to architecture decisions. Used by /architect:blueprint-variants and /architect:recommend-stack to generate informed options.
Invoke this skill to:
Provide 5 constraint dimensions:
{
"budget_monthly": 5000, // dollars/month
"team_size": 3, // engineers
"timeline_weeks": 8, // weeks to launch
"ops_maturity": "startup",
A feasibility analysis object:
{
"constraints": { ... },
"feasibility": {
"overall_score": 0.85, // 0-1.0: can achieve these constraints?
"critical_constraints": [ // hard limits
"budget_monthly: $5000 limits database choice (max PostgreSQL + Redis)"
],
"flexible_constraints": [ // trade-off opportunities
"timeline_weeks: 8 is tight but achievable with baseline tech stack"
]
},
"implied_architecture": {
"deployment": "managed (Vercel, Railway)",
"database": "PostgreSQL managed (Render, Neon, AWS RDS)",
"cache": "Redis managed (Upstash)",
"monitoring": "basic (Datadog free, CloudWatch)"
},
"option_variants": [
{
"name": "baseline",
"matches_constraints": true,
"fit_score": 0.92,
"cost": 4200,
"timeline_weeks": 8,
"ops_burden_hrs_per_week": 4
},
{
"name": "cost-optimized",
"matches_constraints": true,
"fit_score": 0.78,
"cost": 2100,
"timeline_weeks": 6,
"ops_burden_hrs_per_week": 8,
"trade_offs": ["higher latency (2s vs 300ms)", "less predictable costs", "more manual devops"]
}
],
"constraint_conflicts": [
{
"constraints": ["ops_maturity: startup", "compliance_required: HIPAA"],
"conflict": "HIPAA requires enterprise ops (logging, audit trails, multi-region); startup ops maturity can't support this",
"resolution": "hire compliance/ops expert OR relax to SOC 2 instead"
}
],
"recommendations": [
"Budget is your primary constraint. Baseline at $4.2k/mo fits.",
"Team of 3 can execute in 8 weeks with standard tech stack.",
"No compliance constraints = simplifies architecture significantly.",
"Choose Baseline option."
]
}
Typical ranges:
Implications:
$500/mo → use Lambda + SQLite + free monitoring
$2k/mo → managed services (Railway, Render, Neon)
$5k/mo → Kubernetes becomes viable + enterprise services
$20k/mo → multi-cloud, disaster recovery, 24/7 support
Constraint mapping:
Typical team capabilities:
Implications:
1 engineer: Keep it simple. Managed services only. No self-hosted databases.
2 engineers: Full-stack + DevOps split. Can manage 1-2 open-source tools.
3 engineers: Can run Kubernetes cluster, multiple services, complex monitoring.
5+ engineers: Can build internal platforms, custom tooling, research new tech.
Constraint mapping:
Typical development timelines:
Implications:
4 weeks → no time for learning. Use familiar tech only.
8 weeks → baseline timeline for standard stack. Acceptable learning curve.
12 weeks → can onboard new tools, patterns. More experimental stack OK.
16+ weeks → time for complex setup (Kubernetes, multi-region, enterprise hardening).
Constraint mapping:
Levels:
Implications:
Startup: Vercel, Railway, managed databases. Zero self-hosted complexity.
Growing: Some self-hosted (Postgres on own server), but with backups/monitoring.
Mature: Kubernetes, custom infra, complex observability.
Constraint mapping:
Common frameworks:
Implications:
None → standard architecture, any tech stack OK
GDPR → ensure EU data residency, consent flows, no forbidden locations
SOC 2 → add audit logging, role-based access control, annual audit ($20k)
HIPAA → encryption everywhere, multi-region, full audit trail, major cost jump (+$5k/mo)
FedRAMP → government datacenters required, enterprise consulting, $100k+ implementation
Constraint mapping:
Some constraints conflict and can't be satisfied together:
Constraints: budget_monthly: 2000, compliance_required: HIPAA
Problem: HIPAA requires multi-region, 99.99% uptime, 24/7 monitoring
→ Costs minimum $15k/mo, but budget is $2k/mo
Resolution:
Option A: Increase budget to $15k+/mo
Option B: Relax compliance to SOC 2 instead (~$5k/mo)
Option C: Launch without compliance, add later when revenue comes
Constraints: timeline_weeks: 4, ops_maturity: startup
Problem: Startup ops can only use managed services (simple). Takes 4 weeks just to learn tooling.
Resolution:
Option A: Extend timeline to 8 weeks
Option B: Hire experienced ops person (2-week lead time)
Option C: Use even simpler stack (just Vercel + Firebase)
Constraints: team_size: 1, compliance_required: HIPAA, timeline_weeks: 8
Problem: 1 engineer can't implement, test, AND audit for HIPAA. Need 3-4 people.
Resolution:
Option A: Hire additional team members
Option B: Outsource compliance audit ($5k)
Option C: Launch without HIPAA, add compliance after MVP
Calculate 0-1.0 feasibility score for constraint set:
feasibility = 0.0
// Can we actually afford this?
if budget >= calculated_cost:
feasibility += 0.3
elif budget >= calculated_cost * 0.8: // close enough
feasibility += 0.2
else:
flag: "BUDGET INFEASIBLE: need ${amount} more/month"
// Is team large enough?
if team_size >= required_team_size:
feasibility += 0.2
elif team_size >= required_team_size - 1: // close
feasibility += 0.1
else:
flag: "TEAM UNDERSIZED: need ${count} more engineers"
// Is timeline achievable?
if timeline_weeks >= required_weeks:
feasibility += 0.2
elif timeline_weeks >= required_weeks * 0.8:
feasibility += 0.1
else:
flag: "TIMELINE INFEASIBLE: need ${weeks} more weeks"
// Can ops maturity handle this?
if ops_maturity_level >= required_level:
feasibility += 0.15
elif ops_maturity_level >= required_level - 1:
feasibility += 0.075
else:
flag: "OPS MATURITY INSUFFICIENT: hire or use managed services"
// Are compliance requirements supported?
if compliance_frameworks <= team_expertise:
feasibility += 0.15
else:
flag: "COMPLIANCE EXPERTISE LACKING: hire compliance lead"
return min(feasibility, 1.0)
Interpretation:
For each variant (baseline, cost-optimized, etc.), score fit to constraints:
fit_score = 0.0
// Does cost fit budget?
cost_fit = 1.0 - min(1.0, (variant_cost - budget) / budget)
fit_score += cost_fit * 0.25
// Does timeline fit?
timeline_fit = 1.0 - min(1.0, (variant_weeks - timeline_weeks) / timeline_weeks)
fit_score += timeline_fit * 0.25
// Does ops maturity fit?
ops_fit = variant_ops_level <= user_ops_maturity ? 1.0 : 0.5
fit_score += ops_fit * 0.25
// Does it support compliance?
compliance_fit = variant_supports_compliance ? 1.0 : 0.0
fit_score += compliance_fit * 0.25
return fit_score
Example:
Winner: Baseline (0.90 > 0.88 > 0.68)
/architect:blueprint-variants1. Load or ask for constraints
2. Call constraint-solver(constraints)
3. If feasibility < 0.7:
- Show conflicts and suggested changes
- Ask user to adjust constraint
- Recalculate
4. If feasibility >= 0.7:
- Generate all variants
- Score each against constraints
- Rank by fit_score
- Recommend highest-scoring variant
/architect:recommend-stack1. Load constraints
2. For each possible tech stack:
- Calculate cost, complexity, ops burden
- Score fit to constraints
3. Return top 3 stacks ranked by fit_score
4. Show trade-offs of switching stacks
/architect:cost-estimate1. Load constraints (especially budget, timeline)
2. Estimate costs based on constraint targets
3. If estimated cost > budget:
- Suggest cost-optimized variant
- Show what gets cut (latency, scale, features)
{
"budget_monthly": 2000,
"team_size": 2,
"timeline_weeks": 6,
"ops_maturity": "startup",
"compliance_required": []
}
Solver output:
{
"budget_monthly": 8000,
"team_size": 4,
"timeline_weeks": 12,
"ops_maturity": "growing",
"compliance_required": ["SOC2"]
}
Solver output:
{
"budget_monthly": 3000,
"team_size": 1,
"timeline_weeks": 4,
"ops_maturity": "startup",
"compliance_required": ["HIPAA"]
}
Solver output:
blueprint-variants/ — generates options based on constraint analysistech-stack-recommender/ — picks specific tools fitting constraintscost-optimizer/ — helps reduce costs while maintaining quality