AI / LLM PromptingCurated
Instruction Ambiguity Auditor
Run before a prompt goes into production, especially one that feeds a downstream parser.
The prompt
Read this prompt/instruction set I'm about to give an LLM: [paste]. Identify every place a reasonable model could interpret it two different ways, and suggest a rewrite that removes the ambiguity.
Anything in square brackets is meant to be replaced. The builder turns those into fields you can type into.
Tired of re-pasting the same context into every new chat? That is what the Unimatrix MCP server is for.