Agent KB_mediation

Subsystem: mediation

modules/mediation/init.py

• Doc: Mediation layer for LazyOwn cross-module coordination.
• Layer: utility
• Language: py
• Depends on: modules/mediation/chain_mediator.py, modules/mediation/contracts.py, modules/mediation/feedback_mediator.py, modules/mediation/hub.py, modules/mediation/intel_mediator.py, modules/mediation/telemetry_mediator.py
• Imported by: cli/apt_chain_support.py, modules/lazy_intel.py, modules/lazy_loop.py, tests/test_mediation.py

modules/mediation/chain_mediator.py

• Doc: Specialised mediator for the APT chain pipeline.
• Layer: utility
• Language: py
• Symbols:
  • LlmPoolPort (class, line 13) class LlmPoolPort
  • ChainMediator (class, line 48) class ChainMediator
  • complete (method, line 16) def complete(self, prompt, max_tokens, task_id)
  • has_members (method, line 34) def has_members(self)
  • __init__ (method, line 53) def __init__(self)
  • register_llm_port (method, line 57) def register_llm_port(self, port)
  • llm (method, line 65) def llm(self)
  • describe (method, line 73) def describe(self)
• Depends on: modules/lazy_loop.py
• Imported by: modules/mediation/__init__.py, modules/mediation/hub.py, tests/test_guardrails.py

modules/mediation/contracts.py

• Doc: Mediation contracts: the abstractions every participant depends on.
• Layer: utility
• Language: py
• Symbols:
  • MediationDomains (class, line 12) class MediationDomains
  • ActivityLogPort (class, line 35) class ActivityLogPort(Protocol)
  • LlmCompletionPort (class, line 57) class LlmCompletionPort(Protocol)
  • MediatorContract (class, line 84) class MediatorContract(Protocol)
  • MediationHubContract (class, line 95) class MediationHubContract(Protocol)
  • resolve (function, line 103) def resolve(domain)
  • is_known (method, line 23) def is_known(cls, domain)
  • read (method, line 44) def read(self, limit)
  • complete (method, line 60) def complete(self, prompt, max_tokens, task_id)
  • has_members (method, line 78) def has_members(self)
  • describe (method, line 89) def describe(self)
  • mediator_for (method, line 98) def mediator_for(self, domain)
• Depends on: modules/mediation/hub.py
• Imported by: modules/mediation/__init__.py, modules/mediation/hub.py

modules/mediation/feedback_mediator.py

• Doc: Specialised mediator for step-ordering feedback coordination.
• Layer: utility
• Language: py
• Symbols:
  • FeedbackMediator (class, line 13) class FeedbackMediator
  • describe (method, line 18) def describe(self)
  • order_steps (method, line 22) def order_steps(self, steps, stack)
• Depends on: modules/lazy_orchestrator_feedback.py
• Imported by: modules/mediation/__init__.py, modules/mediation/hub.py

modules/mediation/hub.py

• Doc: Thin routing hub for specialised mediators.
• Layer: utility
• Language: py
• Symbols:
  • MediationHub (class, line 19) class MediationHub
  • _build_default (function, line 55) def _build_default()
  • default_hub (function, line 73) def default_hub()
  • get_hub (function, line 78) def get_hub()
  • __init__ (method, line 22) def __init__(self)
  • register (method, line 26) def register(self, mediator)
  • mediator_for (method, line 35) def mediator_for(self, domain)
  • snapshot (method, line 46) def snapshot(self)
• Depends on: modules/mediation/chain_mediator.py, modules/mediation/contracts.py, modules/mediation/feedback_mediator.py, modules/mediation/intel_mediator.py, modules/mediation/telemetry_mediator.py
• Imported by: modules/mediation/__init__.py, modules/mediation/contracts.py, tests/test_mediation.py

modules/mediation/intel_mediator.py

• Doc: specialised mediator for recon-intel coordination.
• Layer: utility
• Language: py
• Symbols:
  • IntelMediator (class, line 13) class IntelMediator
  • __init__ (method, line 18) def __init__(self)
  • register_activity_log_factory (method, line 22) def register_activity_log_factory(self, factory)
  • describe (method, line 32) def describe(self)
  • activity_records (method, line 39) def activity_records(self, task_id, limit)
  • record_routing_decision (method, line 64) def record_routing_decision(self, task_id, intent_id, command, intent_text, target, stack)
  • record_routing_outcome (method, line 97) def record_routing_outcome(self, task_id, intent_id, command, state, facts, findings, stack)
  • mask_configured (method, line 133) def mask_configured(self, text)
  • stack_signature (method, line 149) def stack_signature(self, assets)
• Depends on: modules/lazy_intel.py, modules/lazy_loop.py
• Imported by: modules/mediation/__init__.py, modules/mediation/hub.py

modules/mediation/telemetry_mediator.py

• Doc: Specialised mediator for loop telemetry coordination.
• Layer: utility
• Language: py
• Symbols:
  • TelemetryMediator (class, line 14) class TelemetryMediator
  • describe (method, line 19) def describe(self)
  • publish_event (method, line 23) def publish_event(self, task_id, worker, kind, text, severity)
  • annotate_detection (method, line 43) def annotate_detection(self, command)
  • record_step (method, line 59) def record_step(self, task_id, worker, command, output, detection_prob)
  • swan_available (method, line 76) def swan_available(self)
  • swan_consult (method, line 85) def swan_consult(self, prompt, task_type, phase, timeout)
  • moe_route (method, line 110) def moe_route(self, task_type)
  • rl_select (method, line 126) def rl_select(self, task_type, phase, candidates, deterministic)
  • moe_record_outcome (method, line 151) def moe_record_outcome(self, expert_id, task_type, reward, detection_prob)
  • rl_update (method, line 167) def rl_update(self, task_type, phase, action, reward, candidates, detection_prob)
• Depends on: modules/loop_ai_bridge.py
• Imported by: modules/mediation/__init__.py, modules/mediation/hub.py