Skip to content
Back to Magazine
automation-aiops 3 min read

Agent Incident Response: How to Investigate a Failure When an Agent Was Involved

Does this apply to your company?

Free 30-min AI diagnostic →

Key Takeaways

  • - Intent: what task the agent received.
  • - Context: what information it retrieved or failed to retrieve.
  • - Policy: what rules and permissions applied.
  • - Plan: what path it chose.

Decision

Separate reliable automation from fragile demo before granting it autonomy.

Room

Operations review, architecture, security or platform.

Risk

Adding speed with no observability, rollback, ownership or stop criterion.

Agent prompt: identify guardrails, control points, likely failures and autonomy criteria

Problem

An agent fails, and the postmortem starts with an uncomfortable question: what exactly happened.

The answer rarely fits in a log of prompt and output. The failure can be in retrieval, permissions, tool selection, parameters, retry, hallucination, timeout, validation, or in an external action that seemed successful but wasn’t.

Without a method of investigation, each incident becomes archaeology.

Thesis

Agent Incident Response must become an operational discipline.

It’s not enough to have generic observability. Agents need their own failure taxonomy and a timeline that connects intent, context, decision, tool, and effect.

The goal is not to blame the model. It’s to find which layer failed and what control was missing.

Framework

An agent-related incident must be reconstructed in seven layers:

  • Intent: what task the agent received.
  • Context: what information it retrieved or failed to retrieve.
  • Policy: what rules and permissions applied.
  • Plan: what path it chose.
  • Tools: what calls it made and with what arguments.
  • Outcome: what change occurred in the external system.
  • Verification: how the result was verified.

Mini-case: an agent sends an incorrect email to a client. The problem could be the prompt, outdated CRM, wrong template, lack of human approval, unvalidated tool, or absent verification. Without a timeline, the team discusses opinions. With a timeline, they fix the system.

Measurable signal: percentage of agent-related incidents with a root cause assigned to a specific layer.

Posture: don’t automate what you can’t investigate.

Why It Matters Now

OpenTelemetry already defines conventions for GenAI systems, including spans, events, exceptions, and metrics related to models, agents, and tools. OpenAI recommends safeguards by layer and output validation in agents. MCP documents threats like tool poisoning, confused deputy, and authorization risks.

The infrastructure exists to get started. What’s missing in many companies is the practice: runbooks, labels, owners, and feedback to evaluations.

Anti-Example

“The model made a mistake.”

It’s a lazy explanation. It might be true, but it’s not actionable. A serious incident must identify where the chain broke: context, instruction, tool, permission, verification, or supervision.

Protocol (3 steps)

  1. Create a failure taxonomy. Model, context, tool, permission, integration, criteria, supervision, verification.
  2. Demand a timeline. Each incident must have a sequence of events and tool calls.
  3. Turn causes into evals. Each significant failure must end up as a permanent test case.
LayerQuestionCorrective Action
contextwhat did it knowimprove retrieval
toolwhat did it invokevalidate arguments
permissionwhat could it doreduce scope
outcomewhat changedverify state
evalwould it pass againcreate test

Sources Consulted

Next Step

Reopen the last AI failure you had. Don’t classify it as “hallucination” until you map intent, context, policy, tool, outcome, and verification.


Translated from the Spanish original with AI assistance and reviewed for accuracy. Read the original in Spanish.

incident-response agents observability aiops
Cite this article

Berthelius, V. (2026). “Agent Incident Response: How to Investigate a Failure When an Agent Was Involved”. BRTHLS Magazine. https://www.brthls.com/magazine/agent-incident-response-investigating-failures-en

Fractional CAIO · Free diagnostic

Is your company ready to operate with AI?

30 minutes. No pitch. An honest read on where you are and what to move first.

Book free diagnostic