Remedy's agents use a standardized createAgentHandler architecture, are logged in agent_run_logs and agent_mcp_logs, and support up to 500 events/hour for critical workloads. Provider-agnostic via the RemedyLLM service — each agent uses the optimal model for its task.
Each agent follows the standardized createAgentHandler pattern, with full run logging and sequential dependency build order — earlier agents feed data into later ones.
Pressure/temp/flow anomaly detection with configurable deviation thresholds, risk ranking, and conversational Q&A on operational data.
Well uplift estimation, underperforming well diagnosis, and production opportunity identification across the asset portfolio.
Remaining Useful Life forecasting for pipeline and equipment assets with 30/60/90-day degradation outlook.
API RP 754 Process Safety Event trend analysis, root cause pattern identification, and Tier 1–4 leading indicator monitoring.
Auto-compiles PHMSA Annual Reports, State Production Reports, and EPA Subpart W GHG filings from structured data.
7-rule-type validation engine with AI fix recommendations and auto-issue creation for failed validation rules.
Daily penalty exposure monitoring at $266k/day PHMSA rates with escalating notifications and trend visualization.
Automatic filing cycle generation with collision-proof IDs for recurring regulatory reporting obligations.
Scenario generation for Cyber, Natural Disaster, Pandemic, and Vendor Failure BCP/DR tabletop exercises.
Disaster declaration coordination, communication cascades, and system recovery sequencing at 500 events/hr.
Parallel workstream coordination across containment, evidence preservation, investigation, communications, and recovery.
15-minute CVE/IOC correlation against internal attack surface with CVSS/EPSS weighted prioritization.
CIS Benchmark and DISA STIG baseline auditing with auto-remediation for eligible configuration deviations.
Weighted multi-factor vendor risk assessment: Security 30%, Financial 25%, Compliance 20%, Operational 15%, ESG 10%.
Weekly PSI threshold monitoring for AI/ML model drift with automated alerts when degradation exceeds thresholds.
Context-injected AI assistant with full conversation memory, tool call citations, and feedback tracking across all 18 modules.
Every agent uses the same standardized utility pattern — consistent initialization, error handling, and response formatting across all 16 deployed agents.
All agent runs logged in agent_run_logs and agent_mcp_logs with input/output, latency, token usage, and error tracking.
Per-user and per-tenant configurable thresholds. Critical event agents support up to 500 events/hour with burst handling.
Agents are deployed in dependency order — earlier agents (SCADA, Integrity) feed structured outputs into downstream agents (PSE Analysis, Reporting).
Expanding from 87 core agents to 154 total — covering cross-module orchestration, industry compliance, ESG, AI governance, board reporting, data privacy, and continuous monitoring.
Chain workflows end-to-end across Risk, Audit, Compliance, and Incident modules
HIPAA, NERC CIP, PCI DSS v4.0, SOX, FDA cGMP, ITAR, FedRAMP, BSA/AML, DORA and more
Carbon tracking, TCFD climate risk, supply chain ESG, sustainability report generation
Bias detection, EU AI Act compliance, model performance monitoring, explainability
Board packages, regulatory filings, Audit Committee prep, regulatory exam prep
GDPR orchestration, DSAR automation, cross-border transfers, privacy breach assessment
Always-on control testing, SOD conflict monitoring, configuration compliance, drift detection
Request a live demo and watch Remedy's agents autonomously handle a compliance scenario end-to-end.
Request a Live Demo