Skip to content

SUBSYSTEM_ACTIVATION_STATUS.md

Latest commit

 

History

History
111 lines (93 loc) · 6.63 KB

SUBSYSTEM_ACTIVATION_STATUS.md

File metadata and controls

111 lines (93 loc) · 6.63 KB

Cognitive Subsystem Activation Status

This document records which GödelOS cognitive subsystems have been wired into the active pipeline and which remain genuinely unimplemented or intentionally offline.

Generated for v0.4 – Dormant Module Activation (issue #100).

Active Subsystems (23 / 23 initialise successfully)

# Subsystem Module path Status Notes
1 TypeSystemManager godelOS/core_kr/type_system/manager.py Active Foundational – required by almost everything
2 KnowledgeStoreInterface godelOS/core_kr/knowledge_store/interface.py Active In-memory backend with TRUTHS / BELIEFS / HYPOTHETICAL contexts
3 UnificationEngine godelOS/core_kr/unification_engine/engine.py Active Used by resolution prover and CLP module
4 FormalLogicParser godelOS/core_kr/formal_logic_parser/parser.py Active Parses logic expressions into AST
5 ResolutionProver godelOS/inference_engine/resolution_prover.py Active FOL / propositional resolution
6 ModalTableauProver godelOS/inference_engine/modal_tableau_prover.py Active Was dormant – now registered with InferenceCoordinator
7 CLPModule godelOS/inference_engine/clp_module.py Active Was dormant – constraint logic programming
8 AnalogicalReasoningEngine godelOS/inference_engine/analogical_reasoning_engine.py Active Analogy-based inference
9 InferenceCoordinator godelOS/inference_engine/coordinator.py Active Routes goals to the right prover; all 4 provers registered
10 CachingSystem godelOS/scalability/caching.py Active Memoisation / caching layer
11 NLUPipeline godelOS/nlu_nlg/nlu/pipeline.py Active Text → AST (lexical analysis, semantic interpretation, formalisation)
12 NLGPipeline godelOS/nlu_nlg/nlg/pipeline.py Active AST → Text (content planning, sentence generation, surface realisation)
13 SimulatedEnvironment godelOS/symbol_grounding/simulated_environment.py Active Was dormant – internal world model
14 PerceptualCategorizer godelOS/symbol_grounding/perceptual_categorizer.py Active Was dormant – raw percepts → symbolic vocabulary
15 SymbolGroundingAssociator godelOS/symbol_grounding/symbol_grounding_associator.py Active Was dormant – syntax-to-semantics bridge
16 ActionExecutor godelOS/symbol_grounding/action_executor.py Active Executes actions in the simulated environment
17 InternalStateMonitor godelOS/symbol_grounding/internal_state_monitor.py Active Introspects module state for metacognition
18 ContextEngine godelOS/common_sense/context_engine.py Active Creates / retrieves / merges named contexts
19 CommonSenseContextManager godelOS/common_sense/manager.py Active Was dormant – default reasoning, external KB, contextualised retrieval
20 MetacognitionManager godelOS/metacognition/manager.py Active Monitor → Diagnose → Plan → Modify cycle
21 ILPEngine godelOS/learning_system/ilp_engine.py Active Was dormant – inductive logic programming
22 ExplanationBasedLearner godelOS/learning_system/explanation_based_learner.py Active Was dormant – generalises from proof objects
23 MetaControlRLModule godelOS/learning_system/meta_control_rl_module.py Active Was dormant – RL for meta-level control decisions

Bold entries are the 8 subsystems that were previously dormant (implemented but disconnected) and are now wired into the pipeline.

Pipeline Data Flow

Text Input
    │
    ▼
NLUPipeline  ──────────────────────────┐
    │  (lexical analysis → semantic     │
    │   interpretation → formalisation) │
    ▼                                   │
KnowledgeStoreInterface  ◄─────────────┘
    │  (store / retrieve assertions)
    ▼
InferenceCoordinator
    ├─ ResolutionProver
    ├─ ModalTableauProver  ← dormant, now active
    ├─ CLPModule           ← dormant, now active
    └─ AnalogicalReasoningEngine
    │
    ▼
ContextEngine + CommonSenseContextManager
    │  (context enrichment, default reasoning)
    ▼
NLGPipeline  ─────► Text Output
    (content planning → sentence generation → surface realisation)

Side channels fed during processing:

  • SymbolGroundingAssociator grounds abstract symbols
  • PerceptualCategorizer categorises raw inputs into symbols
  • SimulatedEnvironment supports counterfactual reasoning
  • MetacognitionManager monitors / diagnoses / improves processing
  • ILPEngine induces general rules from examples
  • ExplanationBasedLearner generalises from successful proofs
  • MetaControlRLModule learns optimal strategy selection

Genuinely Unimplemented Modules

The following items are not source-code-complete and therefore cannot be activated:

Module Notes
godelOS/perception/ (directory) Referenced in the issue but does not exist as a separate package. Perception is handled by symbol_grounding/perceptual_categorizer.py.
godelOS/environment/ (directory) Referenced in the issue but does not exist as a separate package. Simulated environment lives in symbol_grounding/simulated_environment.py.
godelOS/inference/modal/enhanced/ An "Enhanced Modal Prover" beyond standard S4/S5 is referenced in the roadmap wiki but has no separate implementation file. The existing ModalTableauProver is activated.

API Endpoints

Endpoint Method Description
/api/status GET System status – now includes cognitive_subsystems with per-subsystem status
/api/system/subsystems GET Dedicated subsystem health endpoint – returns active/total counts and per-subsystem detail

Smoke Tests

tests/test_cognitive_subsystem_activation.py – 26 tests covering:

  • All 23 subsystems initialise without error
  • NLU pipeline processes text
  • Knowledge store has default contexts
  • Inference coordinator has all provers registered (incl. modal & CLP)
  • Symbol grounding components exist
  • Context engine creates / retrieves contexts
  • Common-sense manager is initialised
  • Learning system components exist (ILP, EBL, RL)
  • Metacognition manager can initialise sub-components
  • End-to-end data flow (NLU → KS → Inference → Context → NLG)
  • GödelOSIntegration exposes subsystem status via health endpoint