Ship governance-ready on day one. Each policy defines the agent role, allowed data sources, denied sources, task scoping, and sensitivity ceiling — ready to load and extend.
25 policies across consumer banking, fraud investigation, wealth management, risk & compliance, and operations — built for the strict data boundary requirements of retail and institutional finance.
policies: # Loan underwriter — credit and collateral data; blocked from peer data and internal models - name: loan_underwriter_policy agent_role: loan_underwriter allowed_sources: - account_summaries - credit_scores - loan_history - property_valuations - income_verification denied_sources: - other_customer_data - internal_risk_models - executive_communications allowed_tasks: - credit_decision - collateral_check - risk_assessment denied_tasks: - account_freeze - fraud_flag - product_recommendation max_sensitivity: high
17 policies across clinical operations, compliance & privacy, and revenue cycle — designed for the PHI access controls and audit trail requirements of HIPAA-covered entities.
policies: # Clinical summary agent — EHR and care gap data; blocked from billing and financial records - name: clinical_summary_agent_policy agent_role: clinical_summary_agent allowed_sources: - ehr_summaries - care_gap_data - lab_results - vital_signs denied_sources: - billing_records - insurance_data - audit_logs allowed_tasks: - clinical_summary - care_gap_identification - handoff_preparation denied_tasks: - billing_submission - prescription_write - hipaa_audit max_sensitivity: critical
9 policies across network operations, fraud assurance, and customer experience — covering the CPNI protections and network security requirements unique to telecommunications carriers.
policies: # Network fault agent — topology and alarm data; blocked from CDR and billing data - name: network_fault_agent_policy agent_role: network_fault_agent allowed_sources: - network_topology - alarm_feeds - performance_metrics - change_records denied_sources: - cdr_data - billing_data - customer_profiles allowed_tasks: - fault_detection - root_cause_analysis - ticket_creation denied_tasks: - billing_adjustment - account_suspension - plan_recommendation max_sensitivity: high
9 policies across supply chain, fleet operations, and customs compliance — addressing the cross-border data sovereignty and trade compliance requirements of global logistics operators.
policies: # Demand forecast agent — sales and inventory data; blocked from carrier contracts and driver records - name: demand_forecast_agent_policy agent_role: demand_forecast_agent allowed_sources: - sales_history - inventory_levels - market_signals - supplier_lead_times denied_sources: - carrier_contracts - driver_records - financial_ledgers allowed_tasks: - demand_forecasting - replenishment_planning - supplier_recommendation denied_tasks: - purchase_order_creation - carrier_booking - sanctions_screening max_sensitivity: medium
10 policies across underwriting, claims management, and fraud & compliance — designed for the strict data segregation requirements of P&C, life, and specialty carriers operating under state DOI and Solvency II frameworks.
policies: # Underwriting analyst — risk profile and applicant data; blocked from claims history and competitor pricing - name: underwriting_analyst_policy agent_role: underwriting_analyst allowed_sources: - risk_profile_data - applicant_records - actuarial_tables - property_valuations - credit_reports denied_sources: - claims_history - competitor_pricing - internal_risk_models allowed_tasks: - risk_assessment - policy_quoting - eligibility_determination denied_tasks: - claims_settlement - fraud_investigation max_sensitivity: high
10 policies across merchandising, customer experience, and loss prevention — governing the data boundaries between personalization, pricing, and shrinkage detection in omnichannel retail environments.
policies: # Pricing analyst — cost and market data; blocked from individual customer profiles and payment records - name: pricing_analyst_policy agent_role: pricing_analyst allowed_sources: - cost_data - competitor_pricing - market_benchmarks - margin_reports denied_sources: - individual_customer_profiles - payment_records - legal_contracts allowed_tasks: - price_recommendation - margin_analysis - competitive_benchmarking denied_tasks: - customer_targeting - payment_processing max_sensitivity: medium
10 policies across grid operations, trading compliance, and field safety — enforcing the NERC CIP-mandated boundary between operational technology and energy markets, and isolating field safety workflows from trading data.
policies: # Grid monitoring agent — SCADA feeds and alarm data; blocked from trading positions and customer PII - name: grid_monitoring_agent_policy agent_role: grid_monitoring_agent allowed_sources: - scada_feeds - alarm_feeds - grid_topology - sensor_telemetry denied_sources: - trading_positions - customer_pii - financial_reports allowed_tasks: - grid_monitoring - alarm_triage - fault_detection denied_tasks: - scada_control_commands - trading_execution max_sensitivity: high
10 policies across quality control, supply chain operations, and safety compliance — separating defect detection and SPC data from procurement financials, and isolating OSHA incident records from production and customer data.
policies: # Defect detection agent — sensor and vision data; blocked from supplier contracts and financials - name: defect_detection_agent_policy agent_role: defect_detection_agent allowed_sources: - sensor_data - vision_system_outputs - spc_charts - product_specs denied_sources: - supplier_contracts - financial_ledgers - customer_data allowed_tasks: - defect_detection - quality_flagging - root_cause_initiation denied_tasks: - purchase_order_creation - pricing_decision max_sensitivity: medium
10 policies across property management, transactions, and valuation compliance — enforcing Fair Housing Act data boundaries in tenant screening, maintaining appraiser independence from lender instructions, and separating escrow custody from listing data.
policies: # Tenant screening agent — credit and rental history; blocked from protected class data (Fair Housing Act) - name: tenant_screening_agent_policy agent_role: tenant_screening_agent allowed_sources: - credit_reports - background_check_data - rental_history - income_verification denied_sources: - protected_class_data - medical_records - other_tenant_data allowed_tasks: - credit_check - background_screening - eligibility_determination denied_tasks: - lease_amendment - legal_action_initiation max_sensitivity: high
10 policies across dispensing, clinical review, and regulatory compliance — enforcing DEA schedule boundaries, separating formulary management from individual Rx history, and ensuring pharmacovigilance agents cannot access billing or insurance data.
policies: # Prescription verification agent — Rx data and patient allergies; blocked from billing records and formulary pricing - name: prescription_verification_agent_policy agent_role: prescription_verification_agent allowed_sources: - prescription_records - patient_allergies - drug_database - prescriber_registry denied_sources: - billing_records - formulary_pricing - insurance_contracts allowed_tasks: - prescription_verification - allergy_check - dispense_authorization denied_tasks: - billing_submission - prescription_write max_sensitivity: high
10 policies across citizen services, records management, and procurement compliance — enforcing Privacy Act inter-agency data boundaries, blocking FOIA agents from classified and deliberative records, and isolating source selection data from vendor screening workflows.
policies: # Benefits eligibility agent — income and asset verification; blocked from unrelated agency records (Privacy Act) - name: benefits_eligibility_agent_policy agent_role: benefits_eligibility_agent allowed_sources: - income_records - asset_verification - program_eligibility_data - application_records denied_sources: - unrelated_agency_records - law_enforcement_data - classified_data allowed_tasks: - eligibility_determination - benefit_calculation - application_review denied_tasks: - benefit_disbursement - law_enforcement_referral max_sensitivity: high
5 policies across SaaS platforms — governing AI agents that access code repositories, customer data, telemetry, and internal infrastructure. Enforces least-privilege access across engineering, support, security, and sales engineering roles.
policies: # Software engineer — code and docs access; blocked from customer PII and billing data - name: software_engineer_policy agent_role: software_engineer allowed_sources: - code_repositories - ci_cd_logs - internal_docs - staging_environment denied_sources: - customer_pii - billing_records - production_credentials allowed_tasks: - code_review - test_execution - deployment_staging denied_tasks: - production_deployment - customer_data_export max_sensitivity: high
Point policy_path at the directory. All YAML files are loaded recursively. Industry and category metadata are injected automatically from the directory path — no additional config needed.
Policies are hot-reloaded at runtime. Extend or override any pre-built policy via the REST API without redeploying your agents or restarting services.
from autopil import ContextGuard guard = ContextGuard( policy_path="policies/", # loads all 12 industries recursively audit_db="autopil.db", ) # policies/ directory structure: # policies/ # ├── financial_services/ # │ ├── consumer_banking.yaml # │ ├── wealth.yaml # │ ├── risk_compliance.yaml # │ └── operations.yaml # ├── healthcare/ # │ ├── clinical_operations.yaml # │ ├── compliance_privacy.yaml # │ └── revenue_cycle.yaml # ├── telecom/ # │ ├── network_operations.yaml # │ ├── fraud_assurance.yaml # │ └── customer_experience.yaml # ├── logistics/ # │ ├── supply_chain.yaml # │ ├── fleet_operations.yaml # │ └── customs_compliance.yaml # ├── insurance/ # │ ├── underwriting.yaml # │ ├── claims_management.yaml # │ └── fraud_compliance.yaml # ├── retail/ # │ ├── merchandising.yaml # │ ├── customer_experience.yaml # │ └── loss_prevention.yaml # ├── energy/ # │ ├── grid_operations.yaml # │ ├── trading_compliance.yaml # │ └── field_safety.yaml # ├── manufacturing/ # │ ├── quality_control.yaml # │ ├── supply_chain_ops.yaml # │ └── safety_compliance.yaml # ├── real_estate/ # │ ├── property_management.yaml # │ ├── transactions.yaml # │ └── valuation_compliance.yaml # ├── pharmacy/ # │ ├── dispensing.yaml # │ ├── clinical_review.yaml # │ └── regulatory_compliance.yaml # ├── technology/ # │ └── saas_platform.yaml # └── public_sector/ # ├── citizen_services.yaml # ├── records_management.yaml # └── procurement_compliance.yaml
Start with the policies that match your industry. Extend them via the REST API without redeploying.