@mastra/evals 0.13.8-alpha.1 → 0.13.8

package/dist/scorers/llm/index.js

@@ -220,7 +220,8 @@ function createAnswerRelevancyScorer({
     judge: {
       model,
       instructions: ANSWER_RELEVANCY_AGENT_INSTRUCTIONS
-    }
+    },
+    type: "agent"
   }).preprocess({
     description: "Extract relevant statements from the LLM output",
     outputSchema: extractOutputSchema,
@@ -436,7 +437,8 @@ function createAnswerSimilarityScorer({
     judge: {
       model,
       instructions: ANSWER_SIMILARITY_INSTRUCTIONS
-    }
+    },
+    type: "agent"
   }).preprocess({
     description: "Extract semantic units from output and ground truth",
     outputSchema: extractOutputSchema2,
@@ -692,7 +694,8 @@ function createFaithfulnessScorer({
     judge: {
       model,
       instructions: FAITHFULNESS_AGENT_INSTRUCTIONS
-    }
+    },
+    type: "agent"
   }).preprocess({
     description: "Extract relevant statements from the LLM output",
     outputSchema: z.array(z.string()),
@@ -849,7 +852,8 @@ function createBiasScorer({ model, options }) {
     judge: {
       model,
       instructions: BIAS_AGENT_INSTRUCTIONS
-    }
+    },
+    type: "agent"
   }).preprocess({
     description: "Extract relevant statements from the LLM output",
     outputSchema: z.object({
@@ -1082,7 +1086,8 @@ function createHallucinationScorer({
     judge: {
       model,
       instructions: HALLUCINATION_AGENT_INSTRUCTIONS
-    }
+    },
+    type: "agent"
   }).preprocess({
     description: "Extract all claims from the given output",
     outputSchema: z.object({
@@ -1220,7 +1225,8 @@ function createToxicityScorer({ model, options }) {
     judge: {
       model,
       instructions: TOXICITY_AGENT_INSTRUCTIONS
-    }
+    },
+    type: "agent"
   }).analyze({
     description: "Score the relevance of the statements to the input",
     outputSchema: z.object({ verdicts: z.array(z.object({ verdict: z.string(), reason: z.string() })) }),
@@ -1368,7 +1374,8 @@ function createToolCallAccuracyScorerLLM({ model, availableTools }) {
     judge: {
       model,
       instructions: TOOL_SELECTION_ACCURACY_INSTRUCTIONS
-    }
+    },
+    type: "agent"
   }).preprocess(async ({ run }) => {
     const isInputInvalid = !run.input || !run.input.inputMessages || run.input.inputMessages.length === 0;
     const isOutputInvalid = !run.output || run.output.length === 0;
@@ -1605,7 +1612,8 @@ function createContextRelevanceScorerLLM({
     judge: {
       model,
       instructions: CONTEXT_RELEVANCE_INSTRUCTIONS
-    }
+    },
+    type: "agent"
   }).analyze({
     description: "Analyze the relevance and utility of provided context",
     outputSchema: analyzeOutputSchema3,
@@ -1827,7 +1835,8 @@ function createContextPrecisionScorer({
     judge: {
       model,
       instructions: CONTEXT_PRECISION_AGENT_INSTRUCTIONS
-    }
+    },
+    type: "agent"
   }).analyze({
     description: "Evaluate the relevance of each context piece for generating the expected output",
     outputSchema: contextRelevanceOutputSchema,
@@ -2118,7 +2127,8 @@ function createNoiseSensitivityScorerLLM({
     judge: {
       model,
       instructions: NOISE_SENSITIVITY_INSTRUCTIONS
-    }
+    },
+    type: "agent"
   }).analyze({
     description: "Analyze the impact of noise on agent response quality",
     outputSchema: analyzeOutputSchema4,