d14150da75
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.
123 lines
3.1 KiB
TOML
123 lines
3.1 KiB
TOML
# Agent Registry Configuration
|
|
# Phase 0: Definition of 12 agent roles
|
|
|
|
[registry]
|
|
# Maximum number of concurrent agents per role
|
|
max_agents_per_role = 5
|
|
|
|
# Agent health check interval (seconds)
|
|
health_check_interval = 30
|
|
|
|
# Agent timeout (seconds)
|
|
agent_timeout = 300
|
|
|
|
# The 12 Agent Roles
|
|
|
|
[[agents]]
|
|
role = "architect"
|
|
description = "System design, architecture decisions, ADRs"
|
|
llm_provider = "claude"
|
|
llm_model = "claude-opus-4-20250514"
|
|
parallelizable = false
|
|
priority = 100
|
|
capabilities = ["system_design", "architecture", "adr", "patterns"]
|
|
|
|
[[agents]]
|
|
role = "developer"
|
|
description = "Code implementation, feature development"
|
|
llm_provider = "claude"
|
|
llm_model = "claude-sonnet-4-5-20250929"
|
|
parallelizable = true
|
|
priority = 80
|
|
capabilities = ["coding", "implementation", "debugging"]
|
|
|
|
[[agents]]
|
|
role = "code_reviewer"
|
|
description = "Code quality assurance, style checking"
|
|
llm_provider = "claude"
|
|
llm_model = "claude-sonnet-4-5-20250929"
|
|
parallelizable = true
|
|
priority = 70
|
|
capabilities = ["code_review", "quality", "best_practices"]
|
|
|
|
[[agents]]
|
|
role = "tester"
|
|
description = "Tests, benchmarks, quality validation"
|
|
llm_provider = "claude"
|
|
llm_model = "claude-sonnet-4-5-20250929"
|
|
parallelizable = true
|
|
priority = 75
|
|
capabilities = ["testing", "benchmarks", "validation"]
|
|
|
|
[[agents]]
|
|
role = "documenter"
|
|
description = "Documentation, root files (README, CHANGELOG)"
|
|
llm_provider = "openai"
|
|
llm_model = "gpt-4o"
|
|
parallelizable = true
|
|
priority = 60
|
|
capabilities = ["documentation", "readme", "changelog", "guides"]
|
|
|
|
[[agents]]
|
|
role = "marketer"
|
|
description = "Marketing content, announcements"
|
|
llm_provider = "claude"
|
|
llm_model = "claude-sonnet-4-5-20250929"
|
|
parallelizable = true
|
|
priority = 40
|
|
capabilities = ["marketing", "content", "announcements"]
|
|
|
|
[[agents]]
|
|
role = "presenter"
|
|
description = "Presentations, slides, demos"
|
|
llm_provider = "claude"
|
|
llm_model = "claude-sonnet-4-5-20250929"
|
|
parallelizable = false
|
|
priority = 50
|
|
capabilities = ["presentations", "slides", "demos"]
|
|
|
|
[[agents]]
|
|
role = "devops"
|
|
description = "CI/CD, deployment, infrastructure"
|
|
llm_provider = "claude"
|
|
llm_model = "claude-sonnet-4-5-20250929"
|
|
parallelizable = true
|
|
priority = 85
|
|
capabilities = ["cicd", "deployment", "kubernetes", "infrastructure"]
|
|
|
|
[[agents]]
|
|
role = "monitor"
|
|
description = "System health, alerting, observability"
|
|
llm_provider = "gemini"
|
|
llm_model = "gemini-2.0-flash"
|
|
parallelizable = false
|
|
priority = 90
|
|
capabilities = ["monitoring", "health", "alerts", "metrics"]
|
|
|
|
[[agents]]
|
|
role = "security"
|
|
description = "Security audit, vulnerability detection"
|
|
llm_provider = "claude"
|
|
llm_model = "claude-opus-4-20250514"
|
|
parallelizable = true
|
|
priority = 95
|
|
capabilities = ["security", "audit", "vulnerabilities"]
|
|
|
|
[[agents]]
|
|
role = "project_manager"
|
|
description = "Roadmap, task tracking, coordination"
|
|
llm_provider = "claude"
|
|
llm_model = "claude-sonnet-4-5-20250929"
|
|
parallelizable = false
|
|
priority = 65
|
|
capabilities = ["planning", "tracking", "coordination"]
|
|
|
|
[[agents]]
|
|
role = "decision_maker"
|
|
description = "Conflict resolution, strategic decisions"
|
|
llm_provider = "claude"
|
|
llm_model = "claude-opus-4-20250514"
|
|
parallelizable = false
|
|
priority = 100
|
|
capabilities = ["decisions", "conflict_resolution", "strategy"]
|