The COMPEL Glossary Graph visualizes relationships between framework terminology, showing how concepts interconnect across domains, stages, and pillars. Term nodes cluster by pillar affiliation while cross-references reveal semantic dependencies — for example, how risk appetite connects to control effectiveness, model governance, and assurance requirements. This network representation helps practitioners navigate the framework vocabulary and understand that COMPEL terminology forms a coherent conceptual system rather than isolated definitions.
COMPEL Glossary / prompt
Prompt
The input text, structured message sequence, or multimodal payload provided to a language model at inference time.
What this means in practice
In governance context, a prompt is treated as versioned configuration — not free text — because production behavior depends on the exact prompt content, and auditability requires reproducible prompt versions.
Synonyms
model prompt , LLM prompt , input prompt
See also
- System prompt — Operator-authored instructions that set behavior, persona, and boundaries for a model — distinguished from user prompts because the system prompt carries the operator's policy, while the user prompt carries end-user intent.
- User prompt — End-user input that consumes the system prompt's behavior — the "request" side of an LLM interaction.
- Prompt template — A versioned, parameterized representation of a prompt — with placeholders for user input, retrieved context, and dynamic state — enabling reuse, testing, and audit.
Related articles in the Body of Knowledge
- Prompt Injection and Jailbreak Mitigation
- Prompt Architecture: Templates, Versioning, Injection Defense
- Model, Prompt, and Index Registries
- Prompt Anatomy and the Operator-User Distinction
- Foundational Prompting Patterns
- RAG Prompts and Grounding
- Agentic Prompt Patterns
- Prompt Injection and Safety Boundaries