The Core Thesis
The author uses six named fields to organize prompt requirements. Signal sampling is an analogy; missing a field does not establish that a model will fabricate an answer.
The sinc-prompt specification defines six fields as the author’s approach to organizing task details. It does not guarantee reconstruction of intent. Related: my empirical research.
The Six Bands Defined
Band 0: PERSONA
Who should answer this question? This is not just a job title. It includes the area of expertise, the point of view, the communication style, and how much the reader already knows. When PERSONA is missing, the model defaults to "helpful AI assistant." That is the most generic and least useful voice it has.
Signal example: "You are a principal security engineer at a FAANG company reviewing code for production deployment. You communicate findings as severity-ranked issues with remediation steps. You assume the reader is a senior developer."
Absence cost: The model guesses at expertise level, tone, and reader knowledge. Every guess adds error.
Band 1: CONTEXT
What situation are we in? What happened before this prompt? What is the environment and the project? CONTEXT gives the model the background it needs to give a useful, focused response.
Signal example: "We are in sprint 14 of a 16-sprint migration from a monolith to microservices. 8 of 12 services are deployed. The remaining 4 share a database that cannot be split until all are migrated. Production traffic: 2,400 requests per second."
Absence cost: The model invents a context from average training data. Your real situation gets ignored.
Band 2: DATA
Which facts, numbers and measurements does the task require? Provide reliable inputs and verify how they are used.
Signal example: "Current database: PostgreSQL 14. Rows: 180M in the largest table. Indexes: 23, of which 7 are unused (verified via pg_stat_user_indexes). Average query time: 340ms. P99: 4.7 seconds."
When required inputs are missing, inspect the response for assumptions and unsupported specifics.
Band 3: CONSTRAINTS
What rules, limits, and requirements apply? CONSTRAINTS is the most important band. Nearly half of output quality depends on it. This is the band most people skip entirely.
Signal example: "Never suggest adding hardware. Budget for tooling changes: $0. Maximum acceptable query time: 500ms for p95. All changes must be reversible within 1 hour. Do not modify any table with active writes during business hours (8am-8pm EST). Do not recommend ORM changes."
State relevant limits and evaluate the response against them. Missing constraints do not establish a particular model choice.
Band 4: FORMAT
What shape should the output take? Sections, tables, code blocks, numbered steps, paragraphs, JSON? FORMAT removes a whole dimension of model guessing. It keeps output structure consistent.
Signal example: "Return a numbered list of exactly 5 optimizations. Each item: (1) one-line description, (2) exact SQL or configuration change, (3) expected improvement percentage, (4) risk level (LOW/MEDIUM/HIGH), (5) estimated implementation time."
Absence cost: The model picks a random output shape. Different runs give different formats. Anything that parses that output will break.
Band 5: TASK
What is the actual goal? What do you want the model to produce? TASK is the only band most people include. It has the least impact on quality because it only states the destination. It does not say how to get there.
Signal example: "Identify the 5 highest-impact PostgreSQL performance optimizations for our production database and provide implementation-ready fixes."
Absence cost: Small, since most people do state their task. The problem is that they state only the task.
What Happens When Bands Are Missing
The author uses signal sampling as an analogy for task specification. Six prompt fields organize instructions; the analogy does not guarantee a correct model response. Related: Nyquist-Shannon sampling theorem.
A TASK-only prompt leaves the other five fields unstated. The signal-sampling analogy does not establish a mathematical certainty about the model’s output.
Scroll sideways to see all table columns.
| Fields included | Missing requirements | Output quality | Unsupported claims |
|---|---|---|---|
| 1 of 6 | Inspect task | Not measured | Not measured |
| 2 of 6 | Inspect task | Not measured | Not measured |
| 3 of 6 | Inspect task | Not measured | Not measured |
| 4 of 6 | Inspect task | Not measured | Not measured |
| 5 of 6 | Inspect task | Not measured | Not measured |
| 6 of 6 | Inspect task | Not measured | Not measured |
Aliasing as a Prompt-Specification Analogy
In signal processing, aliasing produces phantom frequencies. These are signals that look real but are artifacts of undersampling. In LLM prompting, aliasing produces phantom facts. These are outputs that look authoritative but come from missing specification bands.
Specification aliasing is the author’s analogy for missing task details. It does not replace factual error analysis or establish that input changes alone fix hallucination.
Band Interaction Effects
Review possible interactions between fields as hypotheses to test. The linked research describes the framework; task-specific results are needed to establish an interaction effect.
- CONSTRAINTS + FORMAT together contribute more than their individual weights suggest. When both are present, the model's probability space collapses on two dimensions at once. This produces a quality gain that multiplies, not just adds.
- PERSONA + CONSTRAINTS interact to set the expertise level at which the model reads your constraints. A "junior developer" persona with "production-grade" constraints produces very different output than a "senior architect" with the same constraints.
- CONTEXT + DATA together remove the model's need to guess the situation. Without both, the model decides what is relevant. With both, relevance is set in advance.
The Optimal Token Allocation
The following 200-230-token allocation is an illustrative planning example. Its percentages and token ranges are assumptions, not measured quality weights or a validated optimum.
Scroll sideways to see all table columns.
| Band | Token Allocation | Target Tokens | |
|---|---|---|---|
| CONSTRAINTS | 42-45% | 85-100 tokens | |
| FORMAT | 25-28% | 50-60 tokens | |
| PERSONA | 10-12% | 20-25 tokens | |
| CONTEXT | 10-12% | 20-25 tokens | |
| DATA | 8-10% | 15-20 tokens | |
| TASK | 3-5% | 5-10 tokens |
Do not assign token shares by a universal formula. Give TASK, CONSTRAINTS and the other fields enough detail for this request, then check for missing requirements, repetition and conflicting instructions.
Practical Application
The sinc-LLM tool organizes prompts into six fields and computes a configured score. Review its output and missing-field suggestions; neither the allocation example nor the score establishes better task performance.
Compare six-field and task-only prompts on representative tasks. A signal-sampling theorem does not establish that one prompt structure produces better model output.
// Production AI Engineering
Build AI systems that hold up in production.
sinc-LLM designs, audits, and stabilises production AI infrastructure: from vendor evaluation and cost accountability to incident controls and MCP architecture.
See what we do →
sinc-LLM