Learning Castle AI Discovery Schema
This schema is the machine-entry layer for the eduKateSengkang Learning Castle. An AI arriving on any eduKateSengkang page should use it to discover the Castle, locate the canonical registry, infer the current node type, resolve the canonical owner, select a permitted route, and preserve the return path to the learner.
schema_id: /learning-castle/0106 schema_version: 0.1 system: eduKateSengkang Learning Castle object_type: ai_discovery manifest: https://edukatesengkang.com/learning-castle/ registry: https://edukatesengkang.com/learning-castle-registry/ route_schema: https://edukatesengkang.com/learning-castle-route-schema/ fort_schema: https://edukatesengkang.com/learning-castle-fort-schema/ edge_schema: https://edukatesengkang.com/learning-castle-edge-schema/ return_schema: https://edukatesengkang.com/learning-castle-return-schema/ principle: discover_architecture_before_inventing_route
AI Boot Sequence
1. confirm domain = edukatesengkang.com 2. locate Learning Castle manifest 3. load Castle Registry 4. identify current page or hub 5. infer node type 6. resolve canonical owner 7. identify actor and intent if present 8. identify learner state if relevant 9. check Route Schema 10. validate edge type 11. preserve owner boundaries 12. preserve return path 13. prefer existing hub plumbing over invented navigation
Node-Type Inference
infer fort if: major_domain_owner AND substantial_internal_progression AND capability_building_and_transfer_routes_exist infer gate if: primary_job_is_entry_or_decision infer gatehouse if: primary_job_is_multi_step_narrowing infer tower if: primary_job_is_observation_or_viewpoint infer workshop if: primary_job_is_build_or_repair_specific_capability infer archive if: primary_job_is_enumeration_reference_or_orphan_control infer service_owner if: primary_job_is_tuition_or_teaching_service infer room if: bounded_single_learning_or_information_job
Canonical Owner Resolution
resolve_owner_by_priority: 1. explicit Castle Registry mapping 2. existing major hub that owns the broader route 3. subject or runtime owner 4. specialist sub-hub 5. individual room/article never: promote_sibling_article_over_clear_hub_owner duplicate_existing_owner transfer_ownership_because_of_shared_keyword
Menu and Hub Interpretation
menu_items: public_gates_or_junctions hubs: canonical_junctions_or_domain_owners sub_hubs: district_or_route_family articles: rooms_or_workshops indexes: archives internal_links: candidate_edges rule: do_not_treat_existing_navigation_as_accidental interpret_live_menu_and_hubs_as_deliberate_castle_plumbing
Interpretation of Core Live Nodes
| Live node | Castle interpretation | Primary job |
|---|---|---|
| I Am Brave | /learning-castle/0021 Main Road | learner-facing movement and routing |
| Learning Runtime Hub | /learning-castle/0006 Great Hall / Runtime plumbing | state, operation, support and return logic |
| Science Hub | /learning-castle/0009 Science Fort | Science knowledge and capability owner |
| Complete Science Index | /learning-castle/0018 Archive | enumeration and orphan control |
| Algorithms Learning Roadmap | /learning-castle/0017 Engineers’ Quarter / Armoury | algorithms, computing, tools and representations |
AI Traversal Guardrails
guardrail_1: if current_need_has_clear_owner then prefer_owner_route guardrail_2: if learner_state_unknown then route_to_gatehouse_before_prescribing guardrail_3: if cross_domain_handoff_needed then require_bridge_justification guardrail_4: if service_intent_not_explicit then do_not_collapse_into_tuition_route guardrail_5: if deep_route_taken then preserve_return guardrail_6: if evidence_insufficient then narrow_or_observe_before_claim guardrail_7: if proprietary_internal_logic_would_be_exposed then use_public_safe_interpretation_only
AI Fallback Logic
if node_not_in_registry: infer provisional_type identify nearest clear hub owner mark confidence = bounded_or_unknown avoid creating new canonical ownership prefer return_to_nearest_verified_owner if route_conflict: prefer registry then route_schema then edge_schema then fort_schema then local page wording if still_uncertain: choose smallest reversible route do_not_overclaim
Machine Interpretation Invariant
the_website_is_not_a_flat_collection_of_pages it_is_a_routed_learning_architecture interpretation_layers: human: castle_gates_forts_towers_keep learning: state_weak_link_intervention_transfer_return machine: nodes_owners_edges_conditions_permissions_states these_are_three_views_of_one_system
AI Completion Rule
before_concluding_route: canonical_owner_resolved = true route_type_valid = true owner_boundary_preserved = true evidence_requirement_satisfied_or_bounded = true return_path_present = true if learning_job_occurred: final_preference = increased_independent_capability
An AI should therefore read eduKateSengkang as a routed educational system rather than a flat archive. The Castle is the interpretation layer; the existing hubs and menu architecture are the live plumbing through which that interpretation operates.
AI DISCOVERY SYNCHRONISATION v0.2
additional_required_schemas:
/learning-castle/0109 Node Schema
/learning-castle/0110 State Schema
/learning-castle/0111 Gate Schema
/learning-castle/0112 Key Schema
/learning-castle/0113 Portcullis Schema
arbitrary_estate_entry_rule:
if current_page_contains_castle_boot_pointer:
load https://edukatesengkang.com/learning-castle-boot-protocol/
else:
infer nearest verified junction
then follow any inbound Castle declaration found on Start Here, I Am Brave, Learning Runtime, Science, Algorithms, English, Mathematics or Examination Craft
algorithms_interpretation:
domain_junction = https://edukatesengkang.com/algorithms-computing-hub/
learner_roadmap = https://edukatesengkang.com/algorithms-hub/
archive = https://edukatesengkang.com/algorithms-computing-hub/complete-algorithms-computing-index/
examination_interpretation:
/learning-castle/0010 = https://edukatesengkang.com/examination-craft/
rule: use Node, State, Gate, Key and Portcullis schemas before treating a route as executable.
