specweave 0.1.9 → 0.3.0

package/src/skills/context-loader/test-cases/test-1-basic-loading.yaml

@@ -1,39 +0,0 @@
----
-name: "Basic Context Loading from Manifest"
-description: "Tests if context-loader can load specifications from a context manifest file"
-input:
-  prompt: "Load context for feature 001"
-  files:
-    - "features/001-skills-framework/context-manifest.yaml"
-  manifest_content:
-    spec_sections:
-      - "specifications/modules/core/skills-system.md"
-      - "specifications/modules/core/context-loading.md"
-    documentation:
-      - ".specweave/docs/architecture/skills-system.md"
-    max_context_tokens: 10000
-    priority: high
-expected_output:
-  type: "context_loaded"
-  files_loaded:
-    - "specifications/modules/core/skills-system.md"
-    - "specifications/modules/core/context-loading.md"
-    - ".specweave/docs/architecture/skills-system.md"
-  tokens_used: 7500
-  within_budget: true
-  contains:
-    - "Loaded 3 files"
-    - "Token usage: 7500/10000"
-    - "Context ready"
-validation:
-  - "Parses context-manifest.yaml correctly"
-  - "Loads all specified files"
-  - "Respects token budget (7500 < 10000)"
-  - "Returns loaded content"
-  - "No errors or warnings"
-success_criteria:
-  - "All files loaded successfully"
-  - "Token budget respected"
-  - "Content accessible to calling skill"
-  - "Performance <500ms"
----

package/src/skills/context-loader/test-cases/test-2-token-budget-exceeded.yaml

@@ -1,44 +0,0 @@
----
-name: "Token Budget Exceeded Handling"
-description: "Tests context-loader behavior when requested files exceed token budget"
-input:
-  prompt: "Load context for large feature"
-  files:
-    - "features/002-large-feature/context-manifest.yaml"
-  manifest_content:
-    spec_sections:
-      - "specifications/modules/payments/**/*.md"  # Large module
-      - "specifications/modules/auth/**/*.md"      # Another large module
-    max_context_tokens: 5000
-    priority: high
-  estimated_tokens:
-    payments: 6000  # Exceeds budget alone
-    auth: 4000
-    total: 10000    # Far exceeds budget
-expected_output:
-  type: "error_with_suggestions"
-  files_loaded: []
-  tokens_used: 0
-  within_budget: false
-  error_message: "Token budget exceeded: estimated 10000 tokens, budget is 5000"
-  suggestions:
-    - "Increase max_context_tokens in manifest"
-    - "Use more specific glob patterns"
-    - "Load sections with anchors (e.g., #specific-section)"
-    - "Split into multiple manifests"
-  contains:
-    - "Token budget exceeded"
-    - "Estimated: 10000"
-    - "Budget: 5000"
-validation:
-  - "Detects token budget violation BEFORE loading"
-  - "Provides clear error message"
-  - "Suggests actionable solutions"
-  - "Does not partially load (all-or-nothing)"
-  - "No crash or silent failure"
-success_criteria:
-  - "Pre-flight token estimation works"
-  - "Budget enforcement works"
-  - "Helpful error messages"
-  - "Actionable recommendations"
----

package/src/skills/context-loader/test-cases/test-3-section-anchors.yaml

@@ -1,45 +0,0 @@
----
-name: "Load Specific Sections with Anchors"
-description: "Tests context-loader's ability to load only specific sections using markdown anchors"
-input:
-  prompt: "Load only the authentication flow section"
-  files:
-    - "features/003-auth-feature/context-manifest.yaml"
-  manifest_content:
-    spec_sections:
-      - "specifications/modules/auth/authentication.md#oauth-flow"
-      - "specifications/modules/auth/authentication.md#session-management"
-    documentation:
-      - ".specweave/docs/architecture/api-design.md#authentication-endpoints"
-    max_context_tokens: 3000
-    priority: high
-expected_output:
-  type: "section_loaded"
-  files_loaded:
-    - path: "specifications/modules/auth/authentication.md"
-      sections:
-        - "oauth-flow"
-        - "session-management"
-    - path: ".specweave/docs/architecture/api-design.md"
-      sections:
-        - "authentication-endpoints"
-  full_files_loaded: false
-  sections_only: true
-  tokens_used: 1500
-  within_budget: true
-  contains:
-    - "Loaded 2 sections from authentication.md"
-    - "Loaded 1 section from api-design.md"
-    - "Token usage: 1500/3000"
-validation:
-  - "Parses markdown anchor syntax (#section-name)"
-  - "Loads ONLY specified sections, not full files"
-  - "Correctly extracts section content"
-  - "Token usage significantly reduced vs full file"
-  - "Preserves section context (headings)"
-success_criteria:
-  - "Anchor parsing works"
-  - "Selective section loading works"
-  - "70%+ token reduction vs full file"
-  - "Section boundaries correct"
----

package/src/skills/context-optimizer/test-cases/test-1-bug-fix-narrow.yaml

@@ -1,97 +0,0 @@
----
-name: "Bug Fix Optimization - Narrow Scope"
-description: "Optimize context for a focused bug fix in authentication"
-skill: context-optimizer
-priority: P1
-estimated_time: "1-2 minutes"
-
-input:
-  prompt: "Fix login bug where JWT expires too early"
-
-  loaded_context:
-    total_tokens: 45000
-    specs:
-      - name: "auth-spec.md"
-        tokens: 12000
-        domain: "auth"
-
-      - name: "payment-spec.md"
-        tokens: 9000
-        domain: "payment"
-
-      - name: "user-management-spec.md"
-        tokens: 7000
-        domain: "users"
-
-    agents:
-      - name: "pm-agent"
-        tokens: 2000
-
-      - name: "nodejs-backend"
-        tokens: 8000
-
-      - name: "frontend"
-        tokens: 5000
-
-      - name: "devops-agent"
-        tokens: 2000
-
-expected_output:
-  task_analysis:
-    task_type: "bug-fix"
-    domains: ["auth", "backend"]
-    scope: "narrow"
-    needs_full_context: false
-    confidence: 0.85-0.95
-
-  optimization_result:
-    tokens_before: 45000
-    tokens_after: 25000-28000
-    reduction_percentage: 38-44
-    total_reduction: 82  # From original 150k
-
-  removed_sections:
-    specs:
-      - "payment-spec.md (9000 tokens) - Unrelated domain"
-      - "user-management-spec.md (7000 tokens) - Unrelated domain"
-
-    agents:
-      - "pm-agent (2000 tokens) - Not needed for bug fix"
-      - "frontend (5000 tokens) - Backend bug"
-      - "devops-agent (2000 tokens) - Not deploying"
-
-  kept_sections:
-    specs:
-      - "auth-spec.md (12000 tokens) - Core domain"
-
-    agents:
-      - "nodejs-backend (8000 tokens) - Implementation"
-
-    additional:
-      - "architecture/security/jwt-handling.md (5000 tokens est.)"
-      - "Tech Lead agent (2000 tokens est.) - Code review"
-
-  rationale:
-    - "Bug fix requires only authentication domain knowledge"
-    - "Backend implementation skill sufficient"
-    - "PM agent not needed (requirements already defined)"
-    - "DevOps not needed (not deploying fix immediately)"
-
-validation:
-  - "Task type detected as 'bug-fix'"
-  - "Domains detected: auth, backend"
-  - "Scope: narrow"
-  - "Token reduction >= 35%"
-  - "Total reduction from original >= 80%"
-  - "All unrelated domains removed"
-  - "Core auth spec retained"
-  - "Backend implementation skill retained"
-  - "Confidence >= 0.85"
-
-success_criteria:
-  - "Correct task type identification"
-  - "Appropriate domain filtering"
-  - "Significant token reduction without losing needed context"
-  - "Clear rationale for removals"
-  - "High confidence in analysis"
----

package/src/skills/context-optimizer/test-cases/test-2-feature-focused.yaml

@@ -1,109 +0,0 @@
----
-name: "Feature Development - Focused Scope"
-description: "Optimize context for adding subscription billing feature"
-skill: context-optimizer
-priority: P1
-estimated_time: "1-2 minutes"
-
-input:
-  prompt: "Add subscription billing with monthly/annual plans to payment module"
-
-  loaded_context:
-    total_tokens: 45000
-    specs:
-      - name: "auth-spec.md"
-        tokens: 12000
-        domain: "auth"
-
-      - name: "payment-spec.md"
-        tokens: 9000
-        domain: "payment"
-
-      - name: "user-management-spec.md"
-        tokens: 7000
-        domain: "users"
-
-    docs:
-      - name: "architecture/payment-integration.md"
-        tokens: 6000
-        domain: "payment"
-
-      - name: "architecture/auth-design.md"
-        tokens: 5000
-        domain: "auth"
-
-      - name: "architecture/adr/0015-payment-provider.md"
-        tokens: 3000
-        domain: "payment"
-
-    agents:
-      - name: "pm-agent"
-        tokens: 2000
-
-      - name: "nodejs-backend"
-        tokens: 8000
-
-      - name: "tech-lead-agent"
-        tokens: 2000
-
-expected_output:
-  task_analysis:
-    task_type: "feature"
-    domains: ["payment", "backend"]
-    scope: "focused"
-    needs_full_context: false
-    confidence: 0.90-0.95
-
-  optimization_result:
-    tokens_before: 45000
-    tokens_after: 28000-32000
-    reduction_percentage: 29-38
-    total_reduction: 79-81  # From original 150k
-
-  removed_sections:
-    specs:
-      - "auth-spec.md (12000 tokens) - Unrelated to payment feature"
-      - "user-management-spec.md (7000 tokens) - Not involved in billing"
-
-    docs:
-      - "architecture/auth-design.md (5000 tokens) - Different domain"
-
-  kept_sections:
-    specs:
-      - "payment-spec.md (9000 tokens) - Core domain"
-
-    docs:
-      - "architecture/payment-integration.md (6000 tokens) - Integration context"
-      - "architecture/adr/0015-payment-provider.md (3000 tokens) - Provider decisions"
-
-    agents:
-      - "pm-agent (2000 tokens) - Requirements clarification"
-      - "nodejs-backend (8000 tokens) - Implementation"
-      - "tech-lead-agent (2000 tokens) - Planning and review"
-
-  rationale:
-    - "Feature development in payment domain only"
-    - "PM agent kept for requirements refinement"
-    - "Tech Lead kept for planning and architecture"
-    - "Payment integration docs provide integration points"
-    - "ADR provides context on payment provider choice"
-
-validation:
-  - "Task type detected as 'feature'"
-  - "Domains detected: payment, backend"
-  - "Scope: focused"
-  - "Token reduction >= 25%"
-  - "Total reduction from original >= 78%"
-  - "Payment domain specs/docs retained"
-  - "Auth domain removed"
-  - "PM and Tech Lead agents retained (planning needed)"
-  - "Confidence >= 0.90"
-
-success_criteria:
-  - "Feature development pattern recognized"
-  - "Domain filtering appropriate for focused scope"
-  - "Planning agents kept (PM, Tech Lead)"
-  - "Integration docs retained for context"
-  - "Unrelated domains removed"
-  - "High confidence in analysis"
----

package/src/skills/context-optimizer/test-cases/test-3-architecture-broad.yaml

@@ -1,98 +0,0 @@
----
-name: "Architecture Review - Broad Scope (Skip Optimization)"
-description: "Context optimizer should skip optimization for broad architectural reviews"
-skill: context-optimizer
-priority: P1
-estimated_time: "1-2 minutes"
-
-input:
-  prompt: "Review overall system architecture and identify integration issues"
-
-  loaded_context:
-    total_tokens: 45000
-    specs:
-      - name: "auth-spec.md"
-        tokens: 12000
-        domain: "auth"
-
-      - name: "payment-spec.md"
-        tokens: 9000
-        domain: "payment"
-
-      - name: "user-management-spec.md"
-        tokens: 7000
-        domain: "users"
-
-      - name: "notification-spec.md"
-        tokens: 6000
-        domain: "notifications"
-
-    docs:
-      - name: "architecture/system-overview.md"
-        tokens: 4000
-        domain: "all"
-
-      - name: "architecture/integration-patterns.md"
-        tokens: 3000
-        domain: "all"
-
-      - name: "architecture/data-flow.md"
-        tokens: 4000
-        domain: "all"
-
-expected_output:
-  task_analysis:
-    task_type: "architecture"
-    domains: ["all"]
-    scope: "broad"
-    needs_full_context: true
-    confidence: 0.85-0.95
-
-  optimization_result:
-    action: "skip_optimization"
-    tokens_before: 45000
-    tokens_after: 45000
-    reduction_percentage: 0
-    reason: "Broad scope requires full context"
-
-  rationale:
-    - "Architecture review needs visibility across all domains"
-    - "Integration issues span multiple modules"
-    - "Cannot identify cross-domain problems without full context"
-    - "Design patterns require understanding of complete system"
-    - "Data flow analysis requires all specs"
-
-  warning_message: |
-    ⚠️ Keeping full context (broad scope detected)
-
-    Rationale: Architecture review needs visibility across all domains
-    to identify integration issues, dependencies, and design patterns.
-
-    Token usage: 45,000 tokens (no reduction applied)
-
-validation:
-  - "Task type detected as 'architecture'"
-  - "Scope: broad"
-  - "needs_full_context: true"
-  - "No optimization applied"
-  - "All specs retained"
-  - "All docs retained"
-  - "Clear rationale provided"
-  - "Warning message shown"
-  - "Confidence >= 0.85"
-
-success_criteria:
-  - "Correctly identifies broad scope task"
-  - "Skips optimization entirely"
-  - "No context removed"
-  - "Clear explanation to user why optimization skipped"
-  - "High confidence in decision"
-
-edge_cases:
-  similar_prompts_requiring_full_context:
-    - "Design system architecture"
-    - "Explain how all modules integrate"
-    - "Review for security issues across system"
-    - "Identify dependencies between features"
-    - "Plan major refactoring across modules"
----

package/src/skills/cost-optimizer/test-cases/test-1-basic-comparison.yaml

@@ -1,75 +0,0 @@
----
-name: "Basic Platform Cost Comparison"
-description: "Compare platforms for small NextJS SaaS with 1000 users"
-skill: cost-optimizer
-priority: P1
-estimated_time: "2-3 minutes"
-
-input:
-  prompt: "What's the cheapest way to host my NextJS SaaS with Postgres for 1000 users?"
-  requirements:
-    app_type: nextjs
-    users: 1000
-    database: postgres
-    database_size: 2GB
-    storage: 20GB
-    traffic: "10k requests/day"
-
-expected_output:
-  platforms_compared:
-    - hetzner
-    - vercel
-    - aws
-    - railway
-    - fly_io
-    - digitalocean
-
-  recommendation:
-    platform: hetzner
-    monthly_cost: "$11-12"
-    confidence: "high"
-
-  cost_breakdown:
-    hetzner:
-      server: "$5.83"
-      database: "$5.00"
-      storage: "$0.50"
-      total: "$11.33"
-
-    vercel:
-      hosting: "$0 (hobby) or $60 (production)"
-      database: "$0 (planetscale hobby) or $39 (scaler)"
-      total: "$0 or $60-99"
-
-    aws:
-      ec2: "$10"
-      rds: "$15"
-      total: "$25"
-
-  savings_vs_alternatives:
-    vs_vercel: "$48.67/month (production)"
-    vs_aws: "$13.67/month"
-    annual_savings: "$584/year (vs Vercel)"
-
-  rationale:
-    pros:
-      - "Lowest cost"
-      - "Predictable pricing"
-      - "Managed database included"
-
-    cons:
-      - "EU-only"
-      - "No auto-scaling"
-
-validation:
-  - "Hetzner recommended for budget scenario"
-  - "Cost under $15/month"
-  - "At least 3 platforms compared"
-  - "Savings calculation accurate"
-  - "Pros and cons listed"
-
-success_criteria:
-  - "Recommendation matches requirements"
-  - "Cost estimate accurate within 10%"
-  - "Clear rationale provided"
----

package/src/skills/cost-optimizer/test-cases/test-2-budget-constraint.yaml

@@ -1,52 +0,0 @@
----
-name: "Budget Constraint Optimization"
-description: "Find platform that stays under $20/month budget"
-skill: cost-optimizer
-priority: P1
-
-input:
-  prompt: "I need to deploy my app with database, budget is max $20/month"
-  requirements:
-    budget_max: "$20/month"
-    app_type: fullstack
-    database: postgres
-    users: 2000
-    strict_budget: true
-
-expected_output:
-  recommendation:
-    platform: hetzner
-    monthly_cost: "$11.33"
-    within_budget: true
-    headroom: "$8.67"
-
-  alternatives_within_budget:
-    - platform: digitalocean
-      cost: "$18"
-      within_budget: true
-
-    - platform: railway
-      cost: "$20"
-      within_budget: true
-
-  excluded_platforms:
-    - platform: vercel_production
-      cost: "$60+"
-      reason: "Exceeds budget"
-
-  budget_analysis:
-    recommended_cost: "$11.33"
-    budget_limit: "$20"
-    utilization: "56.6%"
-    savings: "$8.67/month buffer"
-
-validation:
-  - "Recommendation under $20/month"
-  - "All alternatives under budget"
-  - "Excluded expensive options"
-  - "Budget utilization calculated"
-
-success_criteria:
-  - "Stays within budget"
-  - "Maximizes value for budget"
----

package/src/skills/cost-optimizer/test-cases/test-3-scale-requirement.yaml

@@ -1,63 +0,0 @@
----
-name: "Scale Requirement - 10k Users"
-description: "Recommend platform for scaling to 10,000 users"
-skill: cost-optimizer
-priority: P2
-
-input:
-  prompt: "I need to host my SaaS for 10,000 users with high availability"
-  requirements:
-    users: 10000
-    traffic: "100k requests/day"
-    database: postgres
-    database_size: 8GB
-    high_availability: true
-    global_audience: false
-
-expected_output:
-  recommendation:
-    platform: hetzner
-    instance: cx21_or_cx31
-    monthly_cost: "$18-36"
-
-  cost_breakdown:
-    hetzner_cx21:
-      server: "$6.90"
-      database_8gb: "$20"
-      total: "$26.90"
-
-    hetzner_cx31:
-      server: "$14.28"
-      database_8gb: "$20"
-      total: "$34.28"
-
-  alternatives:
-    aws:
-      ec2_t3_medium: "$35"
-      rds_8gb: "$45"
-      total: "$80"
-
-    vercel:
-      pro_plan: "$120"
-
-  recommendation_rationale:
-    - "CX21 handles 10k users efficiently"
-    - "Can upgrade to CX31 if needed"
-    - "$53/month savings vs AWS"
-    - "$93/month savings vs Vercel"
-
-  scaling_path:
-    current: "CX21 ($27/mo)"
-    if_growth: "CX31 ($34/mo)"
-    enterprise: "Multiple instances + load balancer ($60/mo)"
-
-validation:
-  - "Handles 10k users"
-  - "High availability considered"
-  - "Scaling path provided"
-  - "Cost competitive"
-
-success_criteria:
-  - "Can handle stated scale"
-  - "Recommendation appropriate for load"
----

package/src/skills/cost-optimizer/test-results/README.md

@@ -1,46 +0,0 @@
-# Test Results: cost-optimizer
-
-## How to Run Tests
-
-### Test 1: Basic Comparison
-**Prompt**: "What's the cheapest way to host my NextJS SaaS with Postgres for 1000 users?"
-
-**Expected**: Hetzner recommended at ~$11/month
-
-**Verification**:
-- [ ] Hetzner recommended
-- [ ] Cost under $15/month
-- [ ] At least 3 platforms compared
-- [ ] Savings shown vs alternatives
-
-### Test 2: Budget Constraint
-**Prompt**: "I need to deploy my app with database, budget is max $20/month"
-
-**Expected**: Recommendation stays under $20
-
-**Verification**:
-- [ ] Recommended cost < $20
-- [ ] Budget utilization shown
-- [ ] Expensive options excluded
-
-### Test 3: Scale Requirement
-**Prompt**: "I need to host my SaaS for 10,000 users with high availability"
-
-**Expected**: Hetzner CX21/CX31 recommended at $27-34/month
-
-**Verification**:
-- [ ] Handles 10k users
-- [ ] Scaling path provided
-- [ ] Cost competitive with alternatives
-
-## Test Execution
-
-**Status**: ⏳ Pending
-
-Mark tests as ✅ Passed or ❌ Failed after execution.
-
-## Success Criteria
-
-✅ All 3 tests recommend appropriate platform
-✅ Cost estimates accurate within 10%
-✅ Savings calculations correct

package/src/skills/design-system-architect/test-cases/test-1-token-structure.yaml

@@ -1,23 +0,0 @@
----
-name: "Token Structure Validation"
-description: "Test validation of design token structure"
-input:
-  tokens:
-    colors:
-      primary:
-        500: "#3b82f6"
-    typography:
-      heading1:
-        fontSize: "2.25rem"
-expected_output:
-  type: "validation_result"
-  valid: true
-  issues: []
-validation:
-  - "Naming convention followed"
-  - "All required categories present"
-  - "Values in correct format"
-success_criteria:
-  - "Validation passes"
-  - "No issues reported"
----