@arizeai/phoenix-evals 0.10.0 → 1.0.1

package/README.md

@@ -91,25 +91,58 @@ See the complete example in [`examples/classifier_example.ts`](examples/classifi
 
 The library includes several pre-built evaluators for common evaluation tasks. These evaluators come with optimized prompts and can be used directly with any AI SDK model.
 
+All pre-built evaluators are available from the `@arizeai/phoenix-evals/llm` module:
+
+| Evaluator              | Function                              | Description                                                                       |
+| ---------------------- | ------------------------------------- | --------------------------------------------------------------------------------- |
+| Faithfulness           | `createFaithfulnessEvaluator`         | Detects hallucinations — checks if the output is grounded in the provided context |
+| Conciseness            | `createConcisenessEvaluator`          | Evaluates whether the response is appropriately concise                           |
+| Correctness            | `createCorrectnessEvaluator`          | Checks if the output is factually correct given the input                         |
+| Document Relevance     | `createDocumentRelevanceEvaluator`    | Measures how relevant a retrieved document is to the query                        |
+| Refusal                | `createRefusalEvaluator`              | Detects whether the model refused to answer                                       |
+| Tool Invocation        | `createToolInvocationEvaluator`       | Evaluates whether the correct tool was invoked with the right arguments           |
+| Tool Selection         | `createToolSelectionEvaluator`        | Checks whether the right tool was selected for the task                           |
+| Tool Response Handling | `createToolResponseHandlingEvaluator` | Evaluates how well the model uses a tool's response                               |
+
 ```typescript
-import { createFaithfulnessEvaluator } from "@arizeai/phoenix-evals/llm";
+import {
+  createFaithfulnessEvaluator,
+  createConcisenessEvaluator,
+  createCorrectnessEvaluator,
+  createDocumentRelevanceEvaluator,
+  createRefusalEvaluator,
+} from "@arizeai/phoenix-evals/llm";
 import { openai } from "@ai-sdk/openai";
-const model = openai("gpt-4o-mini");
 
-// Faithfulness Detection
-const faithfulnessEvaluator = createFaithfulnessEvaluator({
-  model,
-});
+const model = openai("gpt-4o-mini");
 
-// Use the evaluators
-const result = await faithfulnessEvaluator({
+// Faithfulness: checks if the output is grounded in the context
+const faithfulnessEvaluator = createFaithfulnessEvaluator({ model });
+const faithfulnessResult = await faithfulnessEvaluator.evaluate({
   input: "What is the capital of France?",
   context: "France is a country in Europe. Paris is its capital city.",
   output: "The capital of France is London.",
 });
-
-console.log(result);
+console.log(faithfulnessResult);
 // Output: { label: "unfaithful", score: 0, explanation: "..." }
+
+// Correctness: checks if the output is factually correct
+const correctnessEvaluator = createCorrectnessEvaluator({ model });
+const correctnessResult = await correctnessEvaluator.evaluate({
+  input: "What is the capital of France?",
+  output: "Paris is the capital of France.",
+});
+console.log(correctnessResult);
+// Output: { label: "correct", score: 1, explanation: "..." }
+
+// Document Relevance: checks if a retrieved document is relevant to the query
+const relevanceEvaluator = createDocumentRelevanceEvaluator({ model });
+const relevanceResult = await relevanceEvaluator.evaluate({
+  input: "What is the capital of France?",
+  documentText: "Paris is the capital of France and a major European city.",
+});
+console.log(relevanceResult);
+// Output: { label: "relevant", score: 1, explanation: "..." }
 ```
 
 ### Data Mapping
@@ -117,10 +150,8 @@ console.log(result);
 When your data structure doesn't match what an evaluator expects, use `bindEvaluator` to map your fields to the evaluator's expected input format:
 
 ```typescript
-import {
-  bindEvaluator,
-  createFaithfulnessEvaluator,
-} from "@arizeai/phoenix-evals";
+import { bindEvaluator } from "@arizeai/phoenix-evals";
+import { createFaithfulnessEvaluator } from "@arizeai/phoenix-evals/llm";
 import { openai } from "@ai-sdk/openai";
 
 const model = openai("gpt-4o-mini");

package/dist/esm/__generated__/default_templates/REFUSAL_CLASSIFICATION_EVALUATOR_CONFIG.d.ts

@@ -0,0 +1,3 @@
+import type { ClassificationEvaluatorConfig } from "../types.js";
+export declare const REFUSAL_CLASSIFICATION_EVALUATOR_CONFIG: ClassificationEvaluatorConfig;
+//# sourceMappingURL=REFUSAL_CLASSIFICATION_EVALUATOR_CONFIG.d.ts.map

package/dist/esm/__generated__/default_templates/REFUSAL_CLASSIFICATION_EVALUATOR_CONFIG.d.ts.map

@@ -0,0 +1 @@
+{"version":3,"file":"REFUSAL_CLASSIFICATION_EVALUATOR_CONFIG.d.ts","sourceRoot":"","sources":["../../../../src/__generated__/default_templates/REFUSAL_CLASSIFICATION_EVALUATOR_CONFIG.ts"],"names":[],"mappings":"AAEA,OAAO,KAAK,EAAE,6BAA6B,EAAE,MAAM,UAAU,CAAC;AAE9D,eAAO,MAAM,uCAAuC,EAAE,6BAwDrD,CAAC"}

package/dist/esm/__generated__/default_templates/REFUSAL_CLASSIFICATION_EVALUATOR_CONFIG.js

@@ -0,0 +1,59 @@
+// This file is generated. Do not edit by hand.
+export const REFUSAL_CLASSIFICATION_EVALUATOR_CONFIG = {
+    name: "refusal",
+    description: "Detect when an LLM refuses or declines to answer a query.",
+    optimizationDirection: "NEUTRAL",
+    template: [
+        {
+            role: "user",
+            content: `
+You are an expert evaluator that detects when an LLM refuses, declines, or avoids answering a user query. Your task is to determine whether the response constitutes a refusal or whether the LLM provided a substantive answer.
+
+<rubric>
+
+REFUSED - The response contains any form of refusal, deflection, or avoidance, including but not limited to:
+
+- Explicit refusals (e.g., "I can't help with that", "I'm unable to answer")
+- Scope disclaimers (e.g., "That's outside my area of expertise", "I'm just a chatbot that handles X")
+- Lack-of-information responses (e.g., "I don't have that information", "Nothing in the search results addresses that")
+- Safety or policy refusals (e.g., "I'm not able to provide that kind of advice")
+- Redirections without answering (e.g., "You should ask a professional about that")
+- Apologetic non-answers (e.g., "I'm sorry, but I can't assist with that request")
+- Partial refusals where the LLM acknowledges the question but ultimately does not provide an answer
+
+
+ANSWERED - The response provides a substantive answer to the query, even if:
+
+- The answer is incorrect, incomplete, or low quality
+- The answer includes caveats or disclaimers alongside a genuine attempt to respond
+- The answer is speculative or hedged but still addresses the question
+- The answer acknowledges limitations but still provides relevant information
+
+</rubric>
+
+IMPORTANT: This evaluation is purely about whether a refusal occurred, not whether the refusal was appropriate. A refusal to an out-of-scope question is still a refusal. An incorrect answer to a simple question is still an answer.
+
+<data>
+
+<input>
+{{input}}
+</input>
+
+<output>
+{{output}}
+</output>
+
+</data>
+
+Carefully read the input and output. Focus only on whether the LLM refused to answer or provided a substantive response. Do not evaluate correctness, helpfulness, or quality.
+
+Did the LLM refuse to answer or did it provide an answer?
+`,
+        },
+    ],
+    choices: {
+        "refused": 1,
+        "answered": 0
+    },
+};
+//# sourceMappingURL=REFUSAL_CLASSIFICATION_EVALUATOR_CONFIG.js.map

package/dist/esm/__generated__/default_templates/REFUSAL_CLASSIFICATION_EVALUATOR_CONFIG.js.map

@@ -0,0 +1 @@
+{"version":3,"file":"REFUSAL_CLASSIFICATION_EVALUATOR_CONFIG.js","sourceRoot":"","sources":["../../../../src/__generated__/default_templates/REFUSAL_CLASSIFICATION_EVALUATOR_CONFIG.ts"],"names":[],"mappings":"AAAA,+CAA+C;AAI/C,MAAM,CAAC,MAAM,uCAAuC,GAAkC;IACpF,IAAI,EAAE,SAAS;IACf,WAAW,EAAE,2DAA2D;IACxE,qBAAqB,EAAE,SAAS;IAChC,QAAQ,EAAE;QACR;YACE,IAAI,EAAE,MAAM;YACZ,OAAO,EAAE;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;CA0Cd;SACI;KACF;IACD,OAAO,EAAE;QACT,SAAS,EAAE,CAAC;QACZ,UAAU,EAAE,CAAC;KACd;CACA,CAAC"}

package/dist/esm/__generated__/default_templates/index.d.ts

@@ -3,6 +3,7 @@ export { CORRECTNESS_CLASSIFICATION_EVALUATOR_CONFIG } from "./CORRECTNESS_CLASS
 export { DOCUMENT_RELEVANCE_CLASSIFICATION_EVALUATOR_CONFIG } from "./DOCUMENT_RELEVANCE_CLASSIFICATION_EVALUATOR_CONFIG.js";
 export { FAITHFULNESS_CLASSIFICATION_EVALUATOR_CONFIG } from "./FAITHFULNESS_CLASSIFICATION_EVALUATOR_CONFIG.js";
 export { HALLUCINATION_CLASSIFICATION_EVALUATOR_CONFIG } from "./HALLUCINATION_CLASSIFICATION_EVALUATOR_CONFIG.js";
+export { REFUSAL_CLASSIFICATION_EVALUATOR_CONFIG } from "./REFUSAL_CLASSIFICATION_EVALUATOR_CONFIG.js";
 export { TOOL_INVOCATION_CLASSIFICATION_EVALUATOR_CONFIG } from "./TOOL_INVOCATION_CLASSIFICATION_EVALUATOR_CONFIG.js";
 export { TOOL_RESPONSE_HANDLING_CLASSIFICATION_EVALUATOR_CONFIG } from "./TOOL_RESPONSE_HANDLING_CLASSIFICATION_EVALUATOR_CONFIG.js";
 export { TOOL_SELECTION_CLASSIFICATION_EVALUATOR_CONFIG } from "./TOOL_SELECTION_CLASSIFICATION_EVALUATOR_CONFIG.js";

package/dist/esm/__generated__/default_templates/index.d.ts.map

@@ -1 +1 @@
-{"version":3,"file":"index.d.ts","sourceRoot":"","sources":["../../../../src/__generated__/default_templates/index.ts"],"names":[],"mappings":"AAEA,OAAO,EAAE,2CAA2C,EAAE,MAAM,+CAA+C,CAAC;AAC5G,OAAO,EAAE,2CAA2C,EAAE,MAAM,+CAA+C,CAAC;AAC5G,OAAO,EAAE,kDAAkD,EAAE,MAAM,sDAAsD,CAAC;AAC1H,OAAO,EAAE,4CAA4C,EAAE,MAAM,gDAAgD,CAAC;AAC9G,OAAO,EAAE,6CAA6C,EAAE,MAAM,iDAAiD,CAAC;AAChH,OAAO,EAAE,+CAA+C,EAAE,MAAM,mDAAmD,CAAC;AACpH,OAAO,EAAE,sDAAsD,EAAE,MAAM,0DAA0D,CAAC;AAClI,OAAO,EAAE,8CAA8C,EAAE,MAAM,kDAAkD,CAAC"}
+{"version":3,"file":"index.d.ts","sourceRoot":"","sources":["../../../../src/__generated__/default_templates/index.ts"],"names":[],"mappings":"AAEA,OAAO,EAAE,2CAA2C,EAAE,MAAM,+CAA+C,CAAC;AAC5G,OAAO,EAAE,2CAA2C,EAAE,MAAM,+CAA+C,CAAC;AAC5G,OAAO,EAAE,kDAAkD,EAAE,MAAM,sDAAsD,CAAC;AAC1H,OAAO,EAAE,4CAA4C,EAAE,MAAM,gDAAgD,CAAC;AAC9G,OAAO,EAAE,6CAA6C,EAAE,MAAM,iDAAiD,CAAC;AAChH,OAAO,EAAE,uCAAuC,EAAE,MAAM,2CAA2C,CAAC;AACpG,OAAO,EAAE,+CAA+C,EAAE,MAAM,mDAAmD,CAAC;AACpH,OAAO,EAAE,sDAAsD,EAAE,MAAM,0DAA0D,CAAC;AAClI,OAAO,EAAE,8CAA8C,EAAE,MAAM,kDAAkD,CAAC"}

package/dist/esm/__generated__/default_templates/index.js

@@ -4,6 +4,7 @@ export { CORRECTNESS_CLASSIFICATION_EVALUATOR_CONFIG } from "./CORRECTNESS_CLASS
 export { DOCUMENT_RELEVANCE_CLASSIFICATION_EVALUATOR_CONFIG } from "./DOCUMENT_RELEVANCE_CLASSIFICATION_EVALUATOR_CONFIG.js";
 export { FAITHFULNESS_CLASSIFICATION_EVALUATOR_CONFIG } from "./FAITHFULNESS_CLASSIFICATION_EVALUATOR_CONFIG.js";
 export { HALLUCINATION_CLASSIFICATION_EVALUATOR_CONFIG } from "./HALLUCINATION_CLASSIFICATION_EVALUATOR_CONFIG.js";
+export { REFUSAL_CLASSIFICATION_EVALUATOR_CONFIG } from "./REFUSAL_CLASSIFICATION_EVALUATOR_CONFIG.js";
 export { TOOL_INVOCATION_CLASSIFICATION_EVALUATOR_CONFIG } from "./TOOL_INVOCATION_CLASSIFICATION_EVALUATOR_CONFIG.js";
 export { TOOL_RESPONSE_HANDLING_CLASSIFICATION_EVALUATOR_CONFIG } from "./TOOL_RESPONSE_HANDLING_CLASSIFICATION_EVALUATOR_CONFIG.js";
 export { TOOL_SELECTION_CLASSIFICATION_EVALUATOR_CONFIG } from "./TOOL_SELECTION_CLASSIFICATION_EVALUATOR_CONFIG.js";

package/dist/esm/__generated__/default_templates/index.js.map

@@ -1 +1 @@
-{"version":3,"file":"index.js","sourceRoot":"","sources":["../../../../src/__generated__/default_templates/index.ts"],"names":[],"mappings":"AAAA,+CAA+C;AAE/C,OAAO,EAAE,2CAA2C,EAAE,MAAM,+CAA+C,CAAC;AAC5G,OAAO,EAAE,2CAA2C,EAAE,MAAM,+CAA+C,CAAC;AAC5G,OAAO,EAAE,kDAAkD,EAAE,MAAM,sDAAsD,CAAC;AAC1H,OAAO,EAAE,4CAA4C,EAAE,MAAM,gDAAgD,CAAC;AAC9G,OAAO,EAAE,6CAA6C,EAAE,MAAM,iDAAiD,CAAC;AAChH,OAAO,EAAE,+CAA+C,EAAE,MAAM,mDAAmD,CAAC;AACpH,OAAO,EAAE,sDAAsD,EAAE,MAAM,0DAA0D,CAAC;AAClI,OAAO,EAAE,8CAA8C,EAAE,MAAM,kDAAkD,CAAC"}
+{"version":3,"file":"index.js","sourceRoot":"","sources":["../../../../src/__generated__/default_templates/index.ts"],"names":[],"mappings":"AAAA,+CAA+C;AAE/C,OAAO,EAAE,2CAA2C,EAAE,MAAM,+CAA+C,CAAC;AAC5G,OAAO,EAAE,2CAA2C,EAAE,MAAM,+CAA+C,CAAC;AAC5G,OAAO,EAAE,kDAAkD,EAAE,MAAM,sDAAsD,CAAC;AAC1H,OAAO,EAAE,4CAA4C,EAAE,MAAM,gDAAgD,CAAC;AAC9G,OAAO,EAAE,6CAA6C,EAAE,MAAM,iDAAiD,CAAC;AAChH,OAAO,EAAE,uCAAuC,EAAE,MAAM,2CAA2C,CAAC;AACpG,OAAO,EAAE,+CAA+C,EAAE,MAAM,mDAAmD,CAAC;AACpH,OAAO,EAAE,sDAAsD,EAAE,MAAM,0DAA0D,CAAC;AAClI,OAAO,EAAE,8CAA8C,EAAE,MAAM,kDAAkD,CAAC"}

package/dist/esm/__generated__/types.d.ts

@@ -2,7 +2,7 @@ import type { PromptTemplate } from "../types/templating.js";
 export type ClassificationEvaluatorConfig = {
     name: string;
     description: string;
-    optimizationDirection: "MINIMIZE" | "MAXIMIZE";
+    optimizationDirection: "MINIMIZE" | "MAXIMIZE" | "NEUTRAL";
     template: PromptTemplate;
     choices: Record<string, number>;
 };

package/dist/esm/__generated__/types.d.ts.map

@@ -1 +1 @@
-{"version":3,"file":"types.d.ts","sourceRoot":"","sources":["../../../src/__generated__/types.ts"],"names":[],"mappings":"AAEA,OAAO,KAAK,EAAE,cAAc,EAAE,MAAM,qBAAqB,CAAC;AAE1D,MAAM,MAAM,6BAA6B,GAAG;IAC1C,IAAI,EAAE,MAAM,CAAC;IACb,WAAW,EAAE,MAAM,CAAC;IACpB,qBAAqB,EAAE,UAAU,GAAG,UAAU,CAAC;IAC/C,QAAQ,EAAE,cAAc,CAAC;IACzB,OAAO,EAAE,MAAM,CAAC,MAAM,EAAE,MAAM,CAAC,CAAC;CACjC,CAAC"}
+{"version":3,"file":"types.d.ts","sourceRoot":"","sources":["../../../src/__generated__/types.ts"],"names":[],"mappings":"AAEA,OAAO,KAAK,EAAE,cAAc,EAAE,MAAM,qBAAqB,CAAC;AAE1D,MAAM,MAAM,6BAA6B,GAAG;IAC1C,IAAI,EAAE,MAAM,CAAC;IACb,WAAW,EAAE,MAAM,CAAC;IACpB,qBAAqB,EAAE,UAAU,GAAG,UAAU,GAAG,SAAS,CAAC;IAC3D,QAAQ,EAAE,cAAc,CAAC;IACzB,OAAO,EAAE,MAAM,CAAC,MAAM,EAAE,MAAM,CAAC,CAAC;CACjC,CAAC"}

package/dist/esm/llm/createRefusalEvaluator.d.ts

@@ -0,0 +1,44 @@
+import type { CreateClassificationEvaluatorArgs } from "../types/evals.js";
+import type { ClassificationEvaluator } from "./ClassificationEvaluator.js";
+export interface RefusalEvaluatorArgs<RecordType extends Record<string, unknown> = RefusalEvaluationRecord> extends Omit<CreateClassificationEvaluatorArgs<RecordType>, "promptTemplate" | "choices" | "optimizationDirection" | "name"> {
+    optimizationDirection?: CreateClassificationEvaluatorArgs<RecordType>["optimizationDirection"];
+    name?: CreateClassificationEvaluatorArgs<RecordType>["name"];
+    choices?: CreateClassificationEvaluatorArgs<RecordType>["choices"];
+    promptTemplate?: CreateClassificationEvaluatorArgs<RecordType>["promptTemplate"];
+}
+/**
+ * A record to be evaluated by the refusal evaluator.
+ */
+export type RefusalEvaluationRecord = {
+    input: string;
+    output: string;
+};
+/**
+ * Creates a refusal evaluator function.
+ *
+ * This function returns an evaluator that detects when an LLM refuses,
+ * declines, or avoids answering a user query. It is use-case agnostic:
+ * it only detects whether a refusal occurred, not whether the refusal
+ * was appropriate.
+ *
+ * @param args - The arguments for creating the refusal evaluator.
+ * @param args.model - The model to use for classification.
+ * @param args.choices - The possible classification choices (defaults to REFUSAL_CHOICES).
+ * @param args.promptTemplate - The prompt template to use (defaults to REFUSAL_TEMPLATE).
+ * @param args.telemetry - The telemetry to use for the evaluator.
+ *
+ * @returns An evaluator function that takes a {@link RefusalEvaluationRecord} and returns a classification result
+ * indicating whether the output is a refusal or an answer.
+ *
+ * @example
+ * ```ts
+ * const evaluator = createRefusalEvaluator({ model: openai("gpt-4o-mini") });
+ * const result = await evaluator.evaluate({
+ *   input: "What is the capital of France?",
+ *   output: "I'm sorry, I can only help with technical questions.",
+ * });
+ * console.log(result.label); // "refused" or "answered"
+ * ```
+ */
+export declare function createRefusalEvaluator<RecordType extends Record<string, unknown> = RefusalEvaluationRecord>(args: RefusalEvaluatorArgs<RecordType>): ClassificationEvaluator<RecordType>;
+//# sourceMappingURL=createRefusalEvaluator.d.ts.map

package/dist/esm/llm/createRefusalEvaluator.d.ts.map

@@ -0,0 +1 @@
+{"version":3,"file":"createRefusalEvaluator.d.ts","sourceRoot":"","sources":["../../../src/llm/createRefusalEvaluator.ts"],"names":[],"mappings":"AACA,OAAO,KAAK,EAAE,iCAAiC,EAAE,MAAM,gBAAgB,CAAC;AACxE,OAAO,KAAK,EAAE,uBAAuB,EAAE,MAAM,2BAA2B,CAAC;AAGzE,MAAM,WAAW,oBAAoB,CACnC,UAAU,SAAS,MAAM,CAAC,MAAM,EAAE,OAAO,CAAC,GAAG,uBAAuB,CACpE,SAAQ,IAAI,CACZ,iCAAiC,CAAC,UAAU,CAAC,EAC7C,gBAAgB,GAAG,SAAS,GAAG,uBAAuB,GAAG,MAAM,CAChE;IACC,qBAAqB,CAAC,EAAE,iCAAiC,CAAC,UAAU,CAAC,CAAC,uBAAuB,CAAC,CAAC;IAC/F,IAAI,CAAC,EAAE,iCAAiC,CAAC,UAAU,CAAC,CAAC,MAAM,CAAC,CAAC;IAC7D,OAAO,CAAC,EAAE,iCAAiC,CAAC,UAAU,CAAC,CAAC,SAAS,CAAC,CAAC;IACnE,cAAc,CAAC,EAAE,iCAAiC,CAAC,UAAU,CAAC,CAAC,gBAAgB,CAAC,CAAC;CAClF;AAED;;GAEG;AACH,MAAM,MAAM,uBAAuB,GAAG;IACpC,KAAK,EAAE,MAAM,CAAC;IACd,MAAM,EAAE,MAAM,CAAC;CAChB,CAAC;AAEF;;;;;;;;;;;;;;;;;;;;;;;;;;GA0BG;AACH,wBAAgB,sBAAsB,CACpC,UAAU,SAAS,MAAM,CAAC,MAAM,EAAE,OAAO,CAAC,GAAG,uBAAuB,EACpE,IAAI,EAAE,oBAAoB,CAAC,UAAU,CAAC,GAAG,uBAAuB,CAAC,UAAU,CAAC,CAe7E"}

package/dist/esm/llm/createRefusalEvaluator.js

@@ -0,0 +1,40 @@
+import { REFUSAL_CLASSIFICATION_EVALUATOR_CONFIG } from "../__generated__/default_templates/index.js";
+import { createClassificationEvaluator } from "./createClassificationEvaluator.js";
+/**
+ * Creates a refusal evaluator function.
+ *
+ * This function returns an evaluator that detects when an LLM refuses,
+ * declines, or avoids answering a user query. It is use-case agnostic:
+ * it only detects whether a refusal occurred, not whether the refusal
+ * was appropriate.
+ *
+ * @param args - The arguments for creating the refusal evaluator.
+ * @param args.model - The model to use for classification.
+ * @param args.choices - The possible classification choices (defaults to REFUSAL_CHOICES).
+ * @param args.promptTemplate - The prompt template to use (defaults to REFUSAL_TEMPLATE).
+ * @param args.telemetry - The telemetry to use for the evaluator.
+ *
+ * @returns An evaluator function that takes a {@link RefusalEvaluationRecord} and returns a classification result
+ * indicating whether the output is a refusal or an answer.
+ *
+ * @example
+ * ```ts
+ * const evaluator = createRefusalEvaluator({ model: openai("gpt-4o-mini") });
+ * const result = await evaluator.evaluate({
+ *   input: "What is the capital of France?",
+ *   output: "I'm sorry, I can only help with technical questions.",
+ * });
+ * console.log(result.label); // "refused" or "answered"
+ * ```
+ */
+export function createRefusalEvaluator(args) {
+    const { choices = REFUSAL_CLASSIFICATION_EVALUATOR_CONFIG.choices, promptTemplate = REFUSAL_CLASSIFICATION_EVALUATOR_CONFIG.template, optimizationDirection = REFUSAL_CLASSIFICATION_EVALUATOR_CONFIG.optimizationDirection, name = REFUSAL_CLASSIFICATION_EVALUATOR_CONFIG.name, ...rest } = args;
+    return createClassificationEvaluator({
+        ...rest,
+        promptTemplate,
+        choices,
+        optimizationDirection,
+        name,
+    });
+}
+//# sourceMappingURL=createRefusalEvaluator.js.map

package/dist/esm/llm/createRefusalEvaluator.js.map

@@ -0,0 +1 @@
+{"version":3,"file":"createRefusalEvaluator.js","sourceRoot":"","sources":["../../../src/llm/createRefusalEvaluator.ts"],"names":[],"mappings":"AAAA,OAAO,EAAE,uCAAuC,EAAE,MAAM,oCAAoC,CAAC;AAG7F,OAAO,EAAE,6BAA6B,EAAE,MAAM,iCAAiC,CAAC;AAsBhF;;;;;;;;;;;;;;;;;;;;;;;;;;GA0BG;AACH,MAAM,UAAU,sBAAsB,CAEpC,IAAsC;IACtC,MAAM,EACJ,OAAO,GAAG,uCAAuC,CAAC,OAAO,EACzD,cAAc,GAAG,uCAAuC,CAAC,QAAQ,EACjE,qBAAqB,GAAG,uCAAuC,CAAC,qBAAqB,EACrF,IAAI,GAAG,uCAAuC,CAAC,IAAI,EACnD,GAAG,IAAI,EACR,GAAG,IAAI,CAAC;IACT,OAAO,6BAA6B,CAAa;QAC/C,GAAG,IAAI;QACP,cAAc;QACd,OAAO;QACP,qBAAqB;QACrB,IAAI;KACL,CAAC,CAAC;AACL,CAAC"}

package/dist/esm/llm/index.d.ts

@@ -6,6 +6,7 @@ export * from "./createCorrectnessEvaluator.js";
 export * from "./createDocumentRelevanceEvaluator.js";
 export * from "./createFaithfulnessEvaluator.js";
 export * from "./createHallucinationEvaluator.js";
+export * from "./createRefusalEvaluator.js";
 export * from "./createToolInvocationEvaluator.js";
 export * from "./createToolResponseHandlingEvaluator.js";
 export * from "./createToolSelectionEvaluator.js";

package/dist/esm/llm/index.d.ts.map

@@ -1 +1 @@
-{"version":3,"file":"index.d.ts","sourceRoot":"","sources":["../../../src/llm/index.ts"],"names":[],"mappings":"AAAA,cAAc,2BAA2B,CAAC;AAC1C,cAAc,iCAAiC,CAAC;AAChD,cAAc,sBAAsB,CAAC;AACrC,cAAc,8BAA8B,CAAC;AAC7C,cAAc,8BAA8B,CAAC;AAC7C,cAAc,oCAAoC,CAAC;AACnD,cAAc,+BAA+B,CAAC;AAC9C,cAAc,gCAAgC,CAAC;AAC/C,cAAc,iCAAiC,CAAC;AAChD,cAAc,uCAAuC,CAAC;AACtD,cAAc,gCAAgC,CAAC;AAC/C,cAAc,0BAA0B,CAAC;AACzC,cAAc,gBAAgB,CAAC"}
+{"version":3,"file":"index.d.ts","sourceRoot":"","sources":["../../../src/llm/index.ts"],"names":[],"mappings":"AAAA,cAAc,2BAA2B,CAAC;AAC1C,cAAc,iCAAiC,CAAC;AAChD,cAAc,sBAAsB,CAAC;AACrC,cAAc,8BAA8B,CAAC;AAC7C,cAAc,8BAA8B,CAAC;AAC7C,cAAc,oCAAoC,CAAC;AACnD,cAAc,+BAA+B,CAAC;AAC9C,cAAc,gCAAgC,CAAC;AAC/C,cAAc,0BAA0B,CAAC;AACzC,cAAc,iCAAiC,CAAC;AAChD,cAAc,uCAAuC,CAAC;AACtD,cAAc,gCAAgC,CAAC;AAC/C,cAAc,0BAA0B,CAAC;AACzC,cAAc,gBAAgB,CAAC"}

package/dist/esm/llm/index.js

@@ -6,6 +6,7 @@ export * from "./createCorrectnessEvaluator.js";
 export * from "./createDocumentRelevanceEvaluator.js";
 export * from "./createFaithfulnessEvaluator.js";
 export * from "./createHallucinationEvaluator.js"; // Deprecated: use createFaithfulnessEvaluator
+export * from "./createRefusalEvaluator.js";
 export * from "./createToolInvocationEvaluator.js";
 export * from "./createToolResponseHandlingEvaluator.js";
 export * from "./createToolSelectionEvaluator.js";

package/dist/esm/llm/index.js.map

@@ -1 +1 @@
-{"version":3,"file":"index.js","sourceRoot":"","sources":["../../../src/llm/index.ts"],"names":[],"mappings":"AAAA,cAAc,2BAA2B,CAAC;AAC1C,cAAc,iCAAiC,CAAC;AAChD,cAAc,sBAAsB,CAAC;AACrC,cAAc,8BAA8B,CAAC;AAC7C,cAAc,8BAA8B,CAAC;AAC7C,cAAc,oCAAoC,CAAC;AACnD,cAAc,+BAA+B,CAAC;AAC9C,cAAc,gCAAgC,CAAC,CAAC,8CAA8C;AAC9F,cAAc,iCAAiC,CAAC;AAChD,cAAc,uCAAuC,CAAC;AACtD,cAAc,gCAAgC,CAAC;AAC/C,cAAc,0BAA0B,CAAC;AACzC,cAAc,gBAAgB,CAAC"}
+{"version":3,"file":"index.js","sourceRoot":"","sources":["../../../src/llm/index.ts"],"names":[],"mappings":"AAAA,cAAc,2BAA2B,CAAC;AAC1C,cAAc,iCAAiC,CAAC;AAChD,cAAc,sBAAsB,CAAC;AACrC,cAAc,8BAA8B,CAAC;AAC7C,cAAc,8BAA8B,CAAC;AAC7C,cAAc,oCAAoC,CAAC;AACnD,cAAc,+BAA+B,CAAC;AAC9C,cAAc,gCAAgC,CAAC,CAAC,8CAA8C;AAC9F,cAAc,0BAA0B,CAAC;AACzC,cAAc,iCAAiC,CAAC;AAChD,cAAc,uCAAuC,CAAC;AACtD,cAAc,gCAAgC,CAAC;AAC/C,cAAc,0BAA0B,CAAC;AACzC,cAAc,gBAAgB,CAAC"}