feat: Phase 5.3 - Multi-Agent Learning Infrastructure
Implement intelligent agent learning from Knowledge Graph execution history
with per-task-type expertise tracking, recency bias, and learning curves.
## Phase 5.3 Implementation
### Learning Infrastructure (✅ Complete)
- LearningProfileService with per-task-type expertise metrics
- TaskTypeExpertise model tracking success_rate, confidence, learning curves
- Recency bias weighting: recent 7 days weighted 3x higher (exponential decay)
- Confidence scoring prevents overfitting: min(1.0, executions / 20)
- Learning curves computed from daily execution windows
### Agent Scoring Service (✅ Complete)
- Unified AgentScore combining SwarmCoordinator + learning profiles
- Scoring formula: 0.3*base + 0.5*expertise + 0.2*confidence
- Rank agents by combined score for intelligent assignment
- Support for recency-biased scoring (recent_success_rate)
- Methods: rank_agents, select_best, rank_agents_with_recency
### KG Integration (✅ Complete)
- KGPersistence::get_executions_for_task_type() - query by agent + task type
- KGPersistence::get_agent_executions() - all executions for agent
- Coordinator::load_learning_profile_from_kg() - core KG→Learning integration
- Coordinator::load_all_learning_profiles() - batch load for multiple agents
- Convert PersistedExecution → ExecutionData for learning calculations
### Agent Assignment Integration (✅ Complete)
- AgentCoordinator uses learning profiles for task assignment
- extract_task_type() infers task type from title/description
- assign_task() scores candidates using AgentScoringService
- Fallback to load-based selection if no learning data available
- Learning profiles stored in coordinator.learning_profiles RwLock
### Profile Adapter Enhancements (✅ Complete)
- create_learning_profile() - initialize empty profiles
- add_task_type_expertise() - set task-type expertise
- update_profile_with_learning() - update swarm profiles from learning
## Files Modified
### vapora-knowledge-graph/src/persistence.rs (+30 lines)
- get_executions_for_task_type(agent_id, task_type, limit)
- get_agent_executions(agent_id, limit)
### vapora-agents/src/coordinator.rs (+100 lines)
- load_learning_profile_from_kg() - core KG integration method
- load_all_learning_profiles() - batch loading for agents
- assign_task() already uses learning-based scoring via AgentScoringService
### Existing Complete Implementation
- vapora-knowledge-graph/src/learning.rs - calculation functions
- vapora-agents/src/learning_profile.rs - data structures and expertise
- vapora-agents/src/scoring.rs - unified scoring service
- vapora-agents/src/profile_adapter.rs - adapter methods
## Tests Passing
- learning_profile: 7 tests ✅
- scoring: 5 tests ✅
- profile_adapter: 6 tests ✅
- coordinator: learning-specific tests ✅
## Data Flow
1. Task arrives → AgentCoordinator::assign_task()
2. Extract task_type from description
3. Query KG for task-type executions (load_learning_profile_from_kg)
4. Calculate expertise with recency bias
5. Score candidates (SwarmCoordinator + learning)
6. Assign to top-scored agent
7. Execution result → KG → Update learning profiles
## Key Design Decisions
✅ Recency bias: 7-day half-life with 3x weight for recent performance
✅ Confidence scoring: min(1.0, total_executions / 20) prevents overfitting
✅ Hierarchical scoring: 30% base load, 50% expertise, 20% confidence
✅ KG query limit: 100 recent executions per task-type for performance
✅ Async loading: load_learning_profile_from_kg supports concurrent loads
## Next: Phase 5.4 - Cost Optimization
Ready to implement budget enforcement and cost-aware provider selection.
2026-01-11 13:03:53 +00:00
|
|
|
[workspace]
|
|
|
|
|
name = "vapora"
|
|
|
|
|
version = "0.2.0"
|
|
|
|
|
description = "Multi-agent multi-IA software development platform"
|
|
|
|
|
|
|
|
|
|
[cluster]
|
|
|
|
|
name = "vapora-cluster"
|
|
|
|
|
cloud_provider = "auto" # auto-detect or specify: aws, gcp, azure, local
|
|
|
|
|
kcl_schema = "kcl/cluster.k"
|
|
|
|
|
min_nodes = 5
|
|
|
|
|
max_nodes = 50
|
|
|
|
|
|
|
|
|
|
[taskservs]
|
|
|
|
|
backend = "taskservs/vapora-backend.toml"
|
|
|
|
|
frontend = "taskservs/vapora-frontend.toml"
|
|
|
|
|
agents = "taskservs/vapora-agents.toml"
|
|
|
|
|
mcp_gateway = "taskservs/vapora-mcp-gateway.toml"
|
|
|
|
|
llm_router = "taskservs/vapora-llm-router.toml"
|
|
|
|
|
|
2026-01-11 21:46:08 +00:00
|
|
|
[storage.surrealdb]
|
|
|
|
|
namespace = "vapora-system"
|
|
|
|
|
replicas = 3
|
|
|
|
|
storage_size = "50Gi"
|
|
|
|
|
storage_class = "rook-ceph"
|
|
|
|
|
|
|
|
|
|
[storage.redis]
|
|
|
|
|
namespace = "vapora-system"
|
|
|
|
|
storage_size = "20Gi"
|
|
|
|
|
storage_class = "ssd"
|
|
|
|
|
|
|
|
|
|
[storage.nats]
|
|
|
|
|
namespace = "vapora-system"
|
|
|
|
|
replicas = 3
|
|
|
|
|
storage_size = "30Gi"
|
|
|
|
|
storage_class = "rook-ceph"
|
feat: Phase 5.3 - Multi-Agent Learning Infrastructure
Implement intelligent agent learning from Knowledge Graph execution history
with per-task-type expertise tracking, recency bias, and learning curves.
## Phase 5.3 Implementation
### Learning Infrastructure (✅ Complete)
- LearningProfileService with per-task-type expertise metrics
- TaskTypeExpertise model tracking success_rate, confidence, learning curves
- Recency bias weighting: recent 7 days weighted 3x higher (exponential decay)
- Confidence scoring prevents overfitting: min(1.0, executions / 20)
- Learning curves computed from daily execution windows
### Agent Scoring Service (✅ Complete)
- Unified AgentScore combining SwarmCoordinator + learning profiles
- Scoring formula: 0.3*base + 0.5*expertise + 0.2*confidence
- Rank agents by combined score for intelligent assignment
- Support for recency-biased scoring (recent_success_rate)
- Methods: rank_agents, select_best, rank_agents_with_recency
### KG Integration (✅ Complete)
- KGPersistence::get_executions_for_task_type() - query by agent + task type
- KGPersistence::get_agent_executions() - all executions for agent
- Coordinator::load_learning_profile_from_kg() - core KG→Learning integration
- Coordinator::load_all_learning_profiles() - batch load for multiple agents
- Convert PersistedExecution → ExecutionData for learning calculations
### Agent Assignment Integration (✅ Complete)
- AgentCoordinator uses learning profiles for task assignment
- extract_task_type() infers task type from title/description
- assign_task() scores candidates using AgentScoringService
- Fallback to load-based selection if no learning data available
- Learning profiles stored in coordinator.learning_profiles RwLock
### Profile Adapter Enhancements (✅ Complete)
- create_learning_profile() - initialize empty profiles
- add_task_type_expertise() - set task-type expertise
- update_profile_with_learning() - update swarm profiles from learning
## Files Modified
### vapora-knowledge-graph/src/persistence.rs (+30 lines)
- get_executions_for_task_type(agent_id, task_type, limit)
- get_agent_executions(agent_id, limit)
### vapora-agents/src/coordinator.rs (+100 lines)
- load_learning_profile_from_kg() - core KG integration method
- load_all_learning_profiles() - batch loading for agents
- assign_task() already uses learning-based scoring via AgentScoringService
### Existing Complete Implementation
- vapora-knowledge-graph/src/learning.rs - calculation functions
- vapora-agents/src/learning_profile.rs - data structures and expertise
- vapora-agents/src/scoring.rs - unified scoring service
- vapora-agents/src/profile_adapter.rs - adapter methods
## Tests Passing
- learning_profile: 7 tests ✅
- scoring: 5 tests ✅
- profile_adapter: 6 tests ✅
- coordinator: learning-specific tests ✅
## Data Flow
1. Task arrives → AgentCoordinator::assign_task()
2. Extract task_type from description
3. Query KG for task-type executions (load_learning_profile_from_kg)
4. Calculate expertise with recency bias
5. Score candidates (SwarmCoordinator + learning)
6. Assign to top-scored agent
7. Execution result → KG → Update learning profiles
## Key Design Decisions
✅ Recency bias: 7-day half-life with 3x weight for recent performance
✅ Confidence scoring: min(1.0, total_executions / 20) prevents overfitting
✅ Hierarchical scoring: 30% base load, 50% expertise, 20% confidence
✅ KG query limit: 100 recent executions per task-type for performance
✅ Async loading: load_learning_profile_from_kg supports concurrent loads
## Next: Phase 5.4 - Cost Optimization
Ready to implement budget enforcement and cost-aware provider selection.
2026-01-11 13:03:53 +00:00
|
|
|
|
|
|
|
|
[monitoring]
|
|
|
|
|
prometheus = true
|
|
|
|
|
grafana = true
|
|
|
|
|
loki = true
|
|
|
|
|
|
|
|
|
|
[security]
|
|
|
|
|
mtls_enabled = true
|
|
|
|
|
network_policies = true
|
|
|
|
|
rbac = true
|
|
|
|
|
vault_integration = true
|
|
|
|
|
|
|
|
|
|
[ingress]
|
|
|
|
|
gateway = "istio"
|
|
|
|
|
domain = "vapora.example.com"
|
|
|
|
|
tls = true
|
|
|
|
|
rate_limit = 1000 # req/sec
|
|
|
|
|
|
|
|
|
|
[scaling]
|
|
|
|
|
enable_hpa = true
|
|
|
|
|
cpu_target = 70
|
|
|
|
|
memory_target = 80
|
|
|
|
|
|
|
|
|
|
[agents]
|
|
|
|
|
# Initial agent pool sizes
|
|
|
|
|
architect = { min = 2, max = 5, model = "claude-opus-4" }
|
|
|
|
|
developer = { min = 5, max = 20, model = "claude-sonnet-4" }
|
|
|
|
|
code_reviewer = { min = 3, max = 10, model = "claude-sonnet-4" }
|
|
|
|
|
tester = { min = 3, max = 10, model = "claude-sonnet-4" }
|
|
|
|
|
documenter = { min = 2, max = 5, model = "gpt-4" }
|
|
|
|
|
marketer = { min = 1, max = 3, model = "claude-sonnet-4" }
|
|
|
|
|
presenter = { min = 1, max = 3, model = "claude-sonnet-4" }
|
|
|
|
|
devops = { min = 2, max = 5, model = "claude-sonnet-4" }
|
|
|
|
|
monitor = { min = 2, max = 5, model = "gemini-pro" }
|
|
|
|
|
security = { min = 2, max = 5, model = "claude-opus-4" }
|
|
|
|
|
project_manager = { min = 1, max = 2, model = "claude-sonnet-4" }
|
|
|
|
|
decision_maker = { min = 1, max = 1, model = "claude-opus-4" }
|
|
|
|
|
orchestrator = { min = 2, max = 5, model = "claude-opus-4" }
|
|
|
|
|
|
|
|
|
|
[llm_router]
|
|
|
|
|
default_fallback_order = ["claude", "openai", "gemini", "ollama"]
|
|
|
|
|
cost_tracking = true
|
|
|
|
|
warn_threshold_daily = 1000 # cents ($10)
|
|
|
|
|
|
|
|
|
|
[environment]
|
|
|
|
|
RUST_LOG = "debug,vapora=trace"
|
|
|
|
|
|
|
|
|
|
[backup]
|
|
|
|
|
enabled = true
|
|
|
|
|
schedule = "daily"
|
|
|
|
|
retention_days = 30
|