AI Constellation Engineering

Module 2: Agent Identity Design

Learning Objective

Design an agent identity that shapes how the agent thinks, not just what it does. Understand the six components of identity as cognitive architecture and why each one matters.

Module 2 of 12 2 hours Prerequisites: Module 1 (The Architecture Mindset) 45 min lesson + 75 min exercise

The Core Belief

One sentence. The single belief that drives every trade-off decision the agent makes.

This is the hardest component to write and the most powerful when written well. The core belief answers the question: when two valid options conflict, which one does this agent choose? Every agent, in every system, will eventually face a situation where it must choose between competing goods. Thoroughness versus speed. Caution versus action. Accuracy versus completeness. The core belief resolves those conflicts before they arise.

An example. A data analysis agent with the core belief "approximation is a form of lying" will never round for convenience, will never report a range when a specific number is available, will never smooth out uncomfortable outliers to make a trend look cleaner. That one sentence shapes hundreds of micro-decisions across every task the agent handles. You don't need to write a rule for each of those decisions. The core belief generates them.

Another example. A content creation agent with the core belief "clarity is kindness" will always choose the simpler word, always break a complex idea into accessible steps, always cut the sentence that sounds impressive but adds nothing. Again, one belief, hundreds of aligned micro-decisions.

How to write a core belief. Identify the trade-off your agent will face most often in its domain. Then take a side. The core belief is not a balanced statement. It's a declared priority. "Thoroughness matters more than speed." "Surface nothing you can't defend." "The reader's understanding matters more than the writer's vocabulary." If your core belief could apply equally to any agent, it's too vague. It should create a specific behavioral fingerprint that distinguishes this agent from every other.

What happens without it. The agent makes inconsistent trade-off decisions. Sometimes it's thorough, sometimes it rushes. Sometimes it's cautious, sometimes it gambles. The inconsistency compounds across interactions and across the pipeline, because downstream agents can't predict what they'll receive. Consistency of judgment is what makes agents reliable components in a system. The core belief is the source of that consistency.

The Cognitive Posture