Skip to main content

Agent Reasoning

Your agent reasons through problems rather than following scripts. Because it builds Deep Context about your codebase, past incidents, and infrastructure, it reasons about your systems — not generic ones. It gathers evidence, selects the right tools, classifies actions by risk, and explains its thinking — all visible in the chat interface.


The reasoning loop

Every message you send goes through the same loop:

Agent reasoning flow: Your question → Understand → Gather context (parallel queries) → Reason (analyze patterns) → Execute safe actions, ask approval for destructive actions, or respond with findings. The loop iterates up to 10 times.

The agent first understands your request and identifies what data is needed. Then it gathers context — querying data sources in parallel (logs, metrics, resource status, deployment history, memory). Next it reasons over gathered data, identifying patterns and forming conclusions. Finally it acts or responds — executing safe actions, requesting approval for risky ones, or presenting findings.

If the problem requires more work, the loop iterates — up to 10 times per turn. After that, your agent asks whether to continue.


Adaptive thinking

For complex problems, your agent shows its reasoning process in the chat. A collapsible Thinking section appears with descriptive titles for each step — like "Exploring Azure health issues" or "Analyzing active alerts" — and elapsed time:

Adaptive thinking showing reasoning steps with titles and elapsed time

Your agent adjusts reasoning depth automatically. A status check gets a quick response. A multi-service outage gets multi-step reasoning with evidence correlation.


Deep Context in reasoning

Your agent doesn't start from scratch — it draws on Deep Context at every step of the reasoning loop. Deep Context has three pillars that shape how your agent thinks:

  • Code analysis — Your agent has already read your repositories, so when an incident hits, it knows where the retry logic lives and what changed recently.
  • Persistent memory — Past investigations, resolution steps, and your uploaded runbooks inform every new conversation.
  • Background intelligence — Codebase analysis, Kusto schema enrichment, and insight generation continuously deepen the agent's understanding.

Memory and knowledge

At the beginning of every conversation, your agent searches memory for relevant context:

What it draws fromHow it improves reasoning
Session insightsLearns from all past conversations — incident investigations, troubleshooting sessions, scheduled task results
Similar symptom patternsRecognizes recurring patterns and jumps to likely causes faster
Your uploaded runbooks and docsFollows your team's procedures instead of generic advice
User preferencesRemembers your environment context and response preferences

The more knowledge you provide — runbooks, architecture docs, team procedures — the more relevant your agent's reasoning becomes. See Memory & Knowledge for how to manage what your agent knows.


Tool selection

Your agent selects tools strategically based on the problem. It starts with all tools registered on the current custom agent, then filters by platform — using only incident tools for the connected incident platform. It further filters by published list to include only tools you've made available, and adjusts as new information emerges during the conversation.

Each custom agent has its own tool set. When your agent delegates to a different custom agent, the available tools change automatically.

For more on what tools are available, see Tools.


Parallel execution

When your agent identifies independent operations — actions that do not depend on each other's output — it issues them simultaneously in a single turn rather than running them one at a time.

For example, if your agent needs to check pod status, service health, and deployment history, it runs all three commands in parallel instead of waiting for each one to complete before starting the next. This reduces the number of reasoning turns and speeds up investigations.

Parallel execution is guided by tool-level prompts that tell the model: "If the commands are independent and can run in parallel, make multiple tool calls in a single message."


Action classification

Your agent classifies every action before executing:

ClassificationBehaviorExamples
SafeExecutes immediatelyQuery logs, check resource status, list deployments
CautiousExecutes with a brief explanationSend emails, post Teams messages
DestructiveRequires your confirmationRestart an app, scale resources, modify configurations

How your agent handles each type depends on your run mode:

Run modeSafeCautiousDestructive
ReviewExecutesExecutesAsks for approval
AutonomousExecutesExecutesExecutes

Conversation management

Several mechanisms keep long conversations productive:

MechanismWhat it does
CompactionWhen conversations get very long, your agent summarizes earlier context while preserving key findings. You can trigger this manually with the /compact command.
Automatic retriesIf a service interruption occurs mid-response, your agent retries transparently.
Error handlingIf a model encounters a temporary issue, your agent displays a user-friendly message ("model is temporarily experiencing issues") instead of a generic internal error.

Cancellation

When you click Stop, your agent immediately halts all operations and prevents retrying the cancelled task. Your next message starts fresh — unless you explicitly modify the cancelled request.


Boundaries

What reasoning doesWhat it does not do
Gathers evidence from multiple sources in parallelGuarantee finding a root cause — evidence may be insufficient
Classifies actions and respects your run modeAuto-remediate without confirmation in Review mode
Explains its thinking step by stepShare investigation methodology across separate agents
Adjusts reasoning depth to problem complexityReplace human judgment for critical decisions

TopicWhat it covers
Root Cause Analysis →Deep investigation with hypothesis trees
Deep Context →How your agent builds understanding of your environment
Run Modes →Review and Autonomous behavior
Memory →How your agent remembers context across conversations
Tools →Built-in and custom tool capabilities
Skills →Domain-specific investigation procedures
Was this page helpful?