Skip to main content

COMPEL Glossary / model-and-prompt-registry

Model and prompt registry

A versioned inventory of models, system prompts, retrieval sources, and guardrails deployed in production.

What this means in practice

Required so that any production behaviour can be traced to a specific configuration and rolled back, and so that audit can reconstruct the state of an LLM feature at any point in time.

Synonyms

prompt registry , LLM feature registry , system prompt registry

See also

  • Evaluation harness — The infrastructure that runs capability, regression, safety, and human-review evaluations on an LLM feature on a defined cadence.
  • System prompt leakage — Extraction of an LLM feature's hidden system prompt and structural instructions through crafted user input.
  • AI incident (for LLMs) — A subtype of the NIST AI RMF MANAGE 1.4 incident concept specific to LLM systems: confident-but-wrong answer, safety bypass, prompt injection success, sensitive data leakage, or policy-violating tool-call execution.

Related articles in the Body of Knowledge