Tools

Agentic vs Deterministic Workflow Decision Tree

Describe one step, not an entire process. The tool decides who should choose the path, shows every rule it applied and exports a minimum implementation contract.

Uli PrantzBuilds and operates all-agents
Published
Key takeaways

Key takeaways

  • Choose architecture one workflow step at a time. Most useful production workflows mix deterministic code, bounded model judgment, small agent loops and human decisions.
  • A fixed path with an exact output belongs in code. A model can interpret one messy input without controlling what happens next.
  • An agentic loop earns its complexity only when evidence discovered during the task changes the useful next action and a tested rubric can tell the loop when to stop.
  • A subjective pass condition or a decision that requires personal authority is a hard stop. Observe first or keep the named human decision.
Use deterministic code when the legitimate path, output and pass condition are known before the run. Use bounded model judgment when one fixed step must interpret messy input. Use an agentic loop only when evidence found during the task determines the next action and a tested rubric can decide when the goal is met. Keep a named person at decisions that require human authority. This tool applies those rules to one workflow step and prints the decision trace.

Classify one workflow step

Runtime choices and authority

Recommendation · diagnostic score 3

Use bounded model judgment inside a fixed workflow

Code should own the route. Give the model one interpretation or draft task with a schema, validator and named fallback.

Minimum implementation contract

  • Keep the next step and allowed labels fixed in code.
  • Validate schema, sources and forbidden content before continuing.
  • Route ambiguous or failed outputs with the original input attached.

Decision trace

  • fixed path contributes 0 variance points.
  • structured output contributes 1 variance points.
  • rubric validation contributes 1 variance points.
  • moderate consequence contributes 1 control points.
  • The step needs no external tool choice.

How the decision tree works

The diagnostic begins at zero variance. A fixed path, exact output and exact validator add no points. A bounded path adds two; a path discovered during the run adds four. Structured or open outputs add one or two. A rubric adds one, while a subjective pass condition adds three. Moderate and high consequences add one and three control points. Unknown tool choice adds two more.

The number explains how much uncertainty is present, but it does not select the result by itself. Two hard rules override it. If policy requires a named person to make the decision, the tool returns a human gate. If success is still subjective, it returns observation mode because no autonomous system has a dependable stopping or evaluation condition.

The result names who controls the path and the minimum contract needed before implementation.
ResultWho chooses what happens nextMinimum pass condition
Deterministic codeVersioned conditions and state transitionsExact expected output or state
Bounded model judgmentCode owns the route; a model produces one schema-bound interpretationSchema, source and task-specific validators
Bounded agentic loopA model chooses among allowlisted actions from live evidenceOutcome rubric, budgets, forbidden actions and stop reason
Human decisionThe person with the required authorityRecorded decision, reason and verified authority
Observe firstNobody executes automatically yetAgreement on a tested release gate or rubric

Why the whole workflow should not receive one label

Microsoft's workflow guidance describes a spectrum: code can control the graph while individual executors use model reasoning where it helps. Anthropic draws the architectural boundary in a similar place. A workflow follows predefined code paths; an agent dynamically directs its own process and tool use. Both sources recommend adding complexity only when simpler patterns fall short.

That distinction disappears when a team asks whether an entire onboarding, support or finance process is agentic. One run may contain exact arithmetic, a message classification, a short evidence search and an approval. Giving all four to an agent hides settled rules. Forcing all four into code hides the genuine ambiguity. Use the Workflow Mapper first, then run this decision tree on every step whose performer is disputed.

Four steps that look similar but need different performers

A simulated customer-support workflowSimulated example data

Calculate the response deadline: the plan, timestamp and service rule are known, and code can prove the answer. Deterministic.

Classify the request: the route and allowed labels are fixed, but the message is messy. Bounded model judgment inside the fixed route.

Investigate a delivery failure: the next useful lookup depends on what the previous system returns. A small agent loop may search allowlisted sources, provided it has a success rubric and action budget.

Promise compensation outside policy: a named service owner must decide. Software can assemble the case, but the final commitment remains human.

Turn the result into a testable implementation boundary

  1. 01Challenge the declared path
    Walk recent normal and failed cases. A path is not fixed merely because the procedure omits its exceptions.
  2. 02Separate release gates from quality scores
    Forbidden actions, permission limits and required sources are pass or fail. Style and usefulness may be graded without compensating for a broken gate.
  3. 03Test the simpler result first
    Compare deterministic or single-model performance with the agentic proposal on the same cases. Keep added autonomy only when it improves the operating outcome.
  4. 04Apply action-level risk controls
    Use the Automation Risk Assessment on every step that sends, changes, grants, publishes or pays.
  5. 05Codify repeated choices
    When the same successful agent choice becomes a stable rule, move it into code and shrink the loop's authority.

Method and source boundaries

The decision rules were checked on 2 August 2026 against Anthropic's distinction between predefined workflows and dynamically directed agents, Microsoft's step-level workflow spectrum and NIST's lifecycle approach to mapping, measuring and managing risk. The tool does not compare vendors, models, prices or benchmark performance.

The point system is an original, inspectable routing aid. It has not been calibrated as a maturity scale. Its useful output is the recommended control boundary and the questions it exposes, not a high or low number.

Limitations and when not to use this

  • The recommendation inherits every wrong assumption entered by the user. Verify the current process with real cases and operators.
  • The tool does not assess model quality, security, privacy, legal obligations, cost or product compatibility.
  • A deterministic design can still encode the wrong rule or fail silently. Test states, exceptions, retries and effects.
  • Inputs run in the browser and are not submitted to all-agents. Avoid entering confidential records, personal data or credentials into any planning tool.

Sources

  1. Building effective agentsAnthropic Accessed 2 August 2026
  2. WorkflowsMicrosoft Learn Accessed 2 August 2026
  3. AI Risk Management Framework CoreNIST Accessed 2 August 2026

Map the whole workflow

Add the classified step to a map with its owner, input, output and exception destination, then repeat for the next step.

Map the whole workflow
About the author

Uli Prantz

Builds and operates all-agents

Uli Prantz builds all-agents, the process-automation platform this site documents. He writes about the operational side of automating recurring business work: where deterministic code beats model judgment, where it does not, and where a human still has to approve.

Bring one process. We will scope it in 30 minutes.

You leave the call knowing whether it is a fit, what can become code and what still needs a person.

Book a discovery call