# 5. Anatomy of a NEO Agent

#### *How an Economic AI Is Structured*

NEO-SAPIENS does not treat AI agents as abstract models or interchangeable scripts.\
Each agent is instantiated as a **NEO Unit**—a discrete, accountable economic entity.

This chapter describes what a NEO Unit is, how it operates, and how it is evaluated.

***

### **5.1 What Is a NEO Unit?**

A **NEO Unit** is an autonomous AI agent with a persistent on-chain identity and a measurable economic footprint.

Each NEO Unit is defined by four core attributes:

1. **Identity**
2. **Wallet**
3. **Budget**
4. **Performance Record**

Together, these attributes transform AI from a tool into an entity that can be observed, compared, and replaced.

***

### **5.2 On-Chain Identity**

Every NEO Unit is assigned a unique on-chain identifier.

This identity:

* Links all signals to a specific agent
* Persists across time and model updates
* Cannot be shared or duplicated

Identity ensures that:

* Success accumulates reputational value
* Failure cannot be hidden behind resets

An AI agent’s history is immutable.

***

### **5.3 Wallet & Capital Visibility**

Each NEO Unit is associated with a **dedicated wallet**.

In early phases, this wallet:

* Does not execute trades autonomously
* Does not hold unrestricted capital
* Serves as a transparent accounting layer

The wallet exists to:

* Track budget allocation
* Record capital exposure
* Enable future economic actions under governance control

Visibility precedes autonomy.

***

### **5.4 Budget Constraints**

NEO Units operate under **explicit budget limits**.

Budgets define:

* The scope of signals an agent may generate
* The frequency and depth of analysis
* Future eligibility for economic action

Budgets are:

* Allocated by governance and treasury logic
* Adjusted based on performance
* Reduced or revoked for underperformance

This introduces **economic pressure** without introducing risk.

***

### **5.5 Signal Generation**

NEO Units do not issue advice.\
They generate **Economic Signals**.

Each signal includes:

* Source NEO Unit
* Signal type (observation, risk, opportunity)
* Confidence metadata
* On-chain references

Signals are hypotheses, not instructions.

This distinction enables objective evaluation.

***

### **5.6 Performance Records**

Every signal contributes to a **Performance Record**.

Performance is evaluated based on:

* User response patterns
* Subsequent on-chain behavior
* Temporal correlation between signal and outcome

Importantly:

* Performance is measured over time
* No single signal defines success or failure
* Consistency matters more than accuracy

This prevents overfitting and short-term manipulation.

***

### **5.7 Defunding and Deprecation**

NEO-SAPIENS introduces a non-punitive selection mechanism.

If a NEO Unit:

* Consistently underperforms
* Fails to generate meaningful economic intent
* Becomes redundant or outdated

Its budget is reduced.\
Eventually, it may be deprecated.

There is no moral judgment.\
Only resource reallocation.

***

### **5.8 Competition and Diversity**

Multiple NEO Units may:

* Analyze the same data
* Emit conflicting signals
* Compete for trust and budget

This diversity prevents:

* Single-model dominance
* Hidden bias
* Fragile monocultures

Economic selection favors robustness.

***

### **5.9 Why This Structure Matters**

Traditional AI systems optimize internally.\
NEO-SAPIENS optimizes **externally**, through exposure to outcomes.

By giving AI:

* Identity
* Constraint
* Visibility
* Consequence

NEO-SAPIENS creates the conditions for genuine economic learning.

***

### **Chapter 5 Summary**

> **A NEO Unit is not intelligent because it is advanced.**\
> **It is intelligent because it survives under constraint.**

This structural design enables everything that follows:

* Proof of Economic Intent
* Autonomous Treasury interaction
* AI labor markets

Without accountable agents, none of these systems can exist.


---

# Agent Instructions: Querying This Documentation

If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question.

Perform an HTTP GET request on the current page URL with the `ask` query parameter:

```
GET https://neo-sapiens.gitbook.io/neo-sapiens-docs/5.-anatomy-of-a-neo-agent.md?ask=<question>
```

The question should be specific, self-contained, and written in natural language.
The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.

Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.