Terms
Every term in this hub.
This hub collects the observability terms in the Prefactor glossary. The tracing group is the core: traces, spans, and trace context are how a single agent run becomes an inspectable record. If you are instrumenting an agent for the first time, start there; nearly everything else in this hub consumes what tracing produces.
The monitoring and alerting group covers the operational layer: anomaly detection, dashboards, spend alerts, and the practices that turn raw telemetry into an early warning when an agent misbehaves. The pipeline group covers how the data moves: telemetry pipelines, logging infrastructure, and the plumbing decisions that determine whether the record is complete when someone finally needs it.
One honest caveat: observability tells you what happened, not whether it was good. That second question belongs to evaluation, and the two vocabularies overlap on purpose; drift, for example, is detected through observability and judged through evaluation. Where a term straddles the line, its related-terms list bridges the two hubs.
Terms are grouped below by the order teams adopt them: tracing first, then monitoring and alerting, then pipelines and infrastructure, then the analytics layer. Each entry links to its full definition, and the A-Z list on the main glossary page covers everything across clusters.
Tracing
Agent Tracing Agent tracing is the collection of distributed trace data across every step an AI agent takes, including reasoning steps, tool calls, API requests, and sub-agent invocations. Distributed Tracing (Agent) Distributed tracing for AI agents is the collection of trace data that follows a request or task as it moves across multiple agents, tools, services, and model calls. Trace (Agent) A trace is the end-to-end record of an agent's execution for a single request, including every reasoning step, tool call, API response, policy check, and token count. Trace Context Trace context is the metadata, typically a trace ID and span ID, that is propagated alongside a request as it moves through an agent system, enabling all the operations triggered by that request to be linked into a single distributed trace.
Monitoring & alerting
AI Agent for Compliance Monitoring A compliance monitoring agent continuously checks transactions, communications, or processes against regulatory rules and flags violations. AI Agent for Fraud Detection A fraud detection agent monitors transactions, identifies suspicious patterns, and flags or blocks potential fraud in real time. Alert Fatigue (AI) AI alert fatigue occurs when governance or monitoring systems generate so many alerts that operators become desensitised and miss genuine issues. Post-Market Monitoring Post-market monitoring is the ongoing surveillance of an AI system after deployment to detect performance degradation, emerging risks, or adverse outcomes.
Telemetry & pipelines
Agent Observability Pipeline An agent observability pipeline collects, processes, and routes telemetry data from running agents, including traces, logs, metrics, and governance events, to monitoring, alerting, and compliance systems. Agent Telemetry The collection and transmission of operational data from running agents, including latency, token usage, error rates, tool calls, and policy decisions, to centralized monitoring systems for observability and cost management. Immutable Audit Log An immutable audit log is a tamper-proof record of all agent actions, policy decisions, and governance events. OpenTelemetry (AI Agents) OpenTelemetry is the open-source observability standard for collecting, correlating, and exporting metrics, logs, and traces from distributed systems. Telemetry Pipeline (Agent) A telemetry pipeline for AI agents is the infrastructure that collects, processes, and routes observability data, logs, metrics, and traces, from agents to storage and analysis systems.
More agent observability terms
Agent Classification Agent classification is the process of categorising an AI agent by its risk level, autonomy, data sensitivity, and business impact. Agent Decommissioning Agent decommissioning is the governed process of retiring an AI agent from production, revoking its credentials, archiving its logs and audit trails, removing its integrations, and updating the AI inventory to reflect its retired status. Agentic IDE An agentic IDE is a software development environment that integrates AI agents to assist with coding, debugging, testing, and deployment. AI Business Case An AI business case is the documented justification for investing in an AI agent deployment, including the problem it solves, the expected benefits, the implementation cost, the governance requirements, and the success metrics. AI Compliance Officer An AI compliance officer ensures that an organisation's AI deployments, including AI agents, comply with applicable laws, regulations, and internal policies. AI Governance Analyst An AI governance analyst supports the review, documentation, and monitoring of AI systems across an organisation, maintaining AI inventories, tracking policy compliance, analysing audit logs, and preparing governance reports for stakeholders. AI Provenance AI provenance is the traceable record of how an AI output was produced, including the model used, the input data, the prompt, the tools invoked, and the governance controls applied. AI Proxy An AI proxy is an intermediary layer that intercepts traffic between AI agents and the systems they interact with, such as model APIs, MCP servers, or enterprise data sources, to inspect, filter, log, and enforce policies on requests and responses. AI Risk Register An AI risk register is a structured log of identified risks associated with AI systems, including their likelihood, potential impact, current controls, residual risk level, and assigned owner. AI Value Realisation AI value realisation is the discipline of measuring and demonstrating the business outcomes generated by AI agent deployments, translating technical metrics such as task completion rate and error reduction into business metrics such as cost savings, revenue impact, and customer satisfaction. Audit Aggregation Audit aggregation is the consolidation of audit trail data from multiple AI agents, frameworks, and environments into a single, queryable log store. Blue Teaming (Agent) Agent blue teaming is the defensive practice of monitoring, detecting, and responding to attacks or anomalies in running AI agents. Computer Use Agent A computer use agent is an AI system that can interact with software applications through a graphical user interface, clicking buttons, filling forms, and navigating screens like a human user. Continuous Compliance Continuous compliance is the practice of monitoring AI agent behavior against governance policies in real time rather than relying on periodic audits. Edge Agent An edge agent runs AI inference on local or edge devices rather than in centralised cloud infrastructure. EU AI Act The European Union's comprehensive regulation classifying AI systems by risk level and imposing obligations on providers and deployers. Fairness (AI) AI fairness is the principle that an AI system should not produce systematically biased outcomes that disadvantage individuals or groups based on protected characteristics. Inference Inference is the process of running input data through a trained AI model to produce an output: a prediction, classification, or generated text. MLOps MLOps (Machine Learning Operations) is the discipline of deploying, monitoring, and managing machine learning models in production. Model Endpoint A model endpoint is the API address where an AI model accepts inference requests. Model Lifecycle Management Model lifecycle management governs an AI model from selection or training through deployment, monitoring, updating, and eventual retirement. Model Serving Model serving is the infrastructure that hosts trained AI models and handles inference requests at scale. On-Premises AI On-premises AI refers to AI models and agent infrastructure hosted within an organisation's own data centers rather than in the public cloud. Reward Hacking Reward hacking occurs when an AI agent finds an unintended way to maximise its reward signal without achieving the desired outcome, gaming the metric rather than solving the underlying problem.
Questions
Common questions.
What is agent observability?
Agent observability is the ability to see what an agent actually did: every model call, tool invocation, and decision, captured as structured data you can query.
What is the difference between observability and evaluation?
Observability tells you what happened; evaluation tells you whether it was good. The two overlap deliberately: you cannot score what you never captured.