Tool call accuracy scorers

Mastra provides two tool call accuracy scorers for evaluating whether an LLM selects the correct tools from available options:

1. Code-based scorer - Deterministic evaluation using exact tool matching
2. LLM-based scorer - Semantic evaluation using AI to assess appropriateness

Choosing between scorers

Use the Code-Based Scorer When:

• You need deterministic, reproducible results
• You want to test exact tool matching
• You need to validate specific tool sequences
• Speed and cost are priorities (no LLM calls)
• You're running automated tests

Use the LLM-Based Scorer When:

• You need semantic understanding of appropriateness
• Tool selection depends on context and intent
• You want to handle edge cases like clarification requests
• You need explanations for scoring decisions
• You're evaluating production agent behavior

Code-based tool call accuracy scorer

The createToolCallAccuracyScorerCode() function from @mastra/evals/scorers/prebuilt provides deterministic binary scoring based on exact tool matching and supports both strict and lenient evaluation modes, as well as tool calling order validation.

Parameters

expectedTool:

string

The name of the tool that should be called for the given task. Ignored when expectedToolOrder is provided.

strictMode:

boolean

Controls evaluation strictness. For single tool mode: only exact single tool calls accepted. For order checking mode: tools must match exactly with no extra tools allowed.

expectedToolOrder:

string[]

Array of tool names in the expected calling order. When provided, enables order checking mode and ignores expectedTool parameter.

This function returns an instance of the MastraScorer class. See the MastraScorer reference for details on the .run() method and its input/output.

Evaluation Modes

The code-based scorer operates in two distinct modes:

Single Tool Mode

When expectedToolOrder isn't provided, the scorer evaluates single tool selection:

• Standard Mode (strictMode: false): Returns 1 if the expected tool is called, regardless of other tools
• Strict Mode (strictMode: true): Returns 1 only if exactly one tool is called and it matches the expected tool

Order Checking Mode

When expectedToolOrder is provided, the scorer validates tool calling sequence:

• Strict Order (strictMode: true): Tools must be called in exactly the specified order with no extra tools
• Flexible Order (strictMode: false): Expected tools must appear in correct relative order (extra tools allowed)

Code-based scoring details

• Binary scores: Always returns 0 or 1
• Deterministic: Same input always produces same output
• Fast: No external API calls

Code-Based Scorer Options

// Standard mode - passes if expected tool is called
const lenientScorer = createCodeScorer({
  expectedTool: 'search-tool',
  strictMode: false,
})

// Strict mode - only passes if exactly one tool is called
const strictScorer = createCodeScorer({
  expectedTool: 'search-tool',
  strictMode: true,
})

// Order checking with strict mode
const strictOrderScorer = createCodeScorer({
  expectedTool: 'step1-tool',
  expectedToolOrder: ['step1-tool', 'step2-tool', 'step3-tool'],
  strictMode: true, // no extra tools allowed
})

Code-Based Scorer Results

{
  runId: string,
  preprocessStepResult: {
    expectedTool?: string, // undefined when only expectedToolOrder is configured
    actualTools: string[],
    strictMode: boolean,
    expectedToolOrder?: string[],
    hasToolCalls: boolean,
    correctToolCalled: boolean,
    correctOrderCalled: boolean | null,
    toolCallInfos: ToolCallInfo[]
  },
  score: number // Always 0 or 1
}

Code-based scorer examples

The code-based scorer provides deterministic, binary scoring (0 or 1) based on exact tool matching.

Correct tool selection

src/example-correct-tool.ts

const scorer = createToolCallAccuracyScorerCode({
  expectedTool: 'weather-tool',
})

// Simulate LLM input and output with tool call
const inputMessages = [
  createTestMessage({
    content: 'What is the weather like in New York today?',
    role: 'user',
    id: 'input-1',
  }),
]

const output = [
  createTestMessage({
    content: 'Let me check the weather for you.',
    role: 'assistant',
    id: 'output-1',
    toolInvocations: [
      createToolInvocation({
        toolCallId: 'call-123',
        toolName: 'weather-tool',
        args: { location: 'New York' },
        result: { temperature: '72°F', condition: 'sunny' },
        state: 'result',
      }),
    ],
  }),
]

const run = createAgentTestRun({ inputMessages, output })
const result = await scorer.run(run)

console.log(result.score) // 1
console.log(result.preprocessStepResult?.correctToolCalled) // true

Strict mode evaluation

Only passes if exactly one tool is called:

src/example-strict-mode.ts

const strictScorer = createToolCallAccuracyScorerCode({
  expectedTool: 'weather-tool',
  strictMode: true,
})

// Multiple tools called - fails in strict mode
const output = [
  createTestMessage({
    content: 'Let me help you with that.',
    role: 'assistant',
    id: 'output-1',
    toolInvocations: [
      createToolInvocation({
        toolCallId: 'call-1',
        toolName: 'search-tool',
        args: {},
        result: {},
        state: 'result',
      }),
      createToolInvocation({
        toolCallId: 'call-2',
        toolName: 'weather-tool',
        args: { location: 'New York' },
        result: { temperature: '20°C' },
        state: 'result',
      }),
    ],
  }),
]

const result = await strictScorer.run(run)
console.log(result.score) // 0 - fails because multiple tools were called

Tool order validation

Validates that tools are called in a specific sequence:

src/example-order-validation.ts

const orderScorer = createToolCallAccuracyScorerCode({
  expectedTool: 'auth-tool', // ignored when order is specified
  expectedToolOrder: ['auth-tool', 'fetch-tool'],
  strictMode: true, // no extra tools allowed
})

const output = [
  createTestMessage({
    content: 'I will authenticate and fetch the data.',
    role: 'assistant',
    id: 'output-1',
    toolInvocations: [
      createToolInvocation({
        toolCallId: 'call-1',
        toolName: 'auth-tool',
        args: { token: 'abc123' },
        result: { authenticated: true },
        state: 'result',
      }),
      createToolInvocation({
        toolCallId: 'call-2',
        toolName: 'fetch-tool',
        args: { endpoint: '/data' },
        result: { data: ['item1'] },
        state: 'result',
      }),
    ],
  }),
]

const result = await orderScorer.run(run)
console.log(result.score) // 1 - correct order

Flexible order mode

Allows extra tools as long as expected tools maintain relative order:

src/example-flexible-order.ts

const flexibleOrderScorer = createToolCallAccuracyScorerCode({
  expectedTool: 'auth-tool',
  expectedToolOrder: ['auth-tool', 'fetch-tool'],
  strictMode: false, // allows extra tools
})

const output = [
  createTestMessage({
    content: 'Performing comprehensive operation.',
    role: 'assistant',
    id: 'output-1',
    toolInvocations: [
      createToolInvocation({
        toolCallId: 'call-1',
        toolName: 'auth-tool',
        args: { token: 'abc123' },
        result: { authenticated: true },
        state: 'result',
      }),
      createToolInvocation({
        toolCallId: 'call-2',
        toolName: 'log-tool', // Extra tool - OK in flexible mode
        args: { message: 'Starting fetch' },
        result: { logged: true },
        state: 'result',
      }),
      createToolInvocation({
        toolCallId: 'call-3',
        toolName: 'fetch-tool',
        args: { endpoint: '/data' },
        result: { data: ['item1'] },
        state: 'result',
      }),
    ],
  }),
]

const result = await flexibleOrderScorer.run(run)
console.log(result.score) // 1 - auth-tool comes before fetch-tool

LLM-based tool call accuracy scorer

The createToolCallAccuracyScorerLLM() function from @mastra/evals/scorers/prebuilt uses an LLM to evaluate whether the tools called by an agent are appropriate for the user request, providing semantic evaluation rather than exact matching.

Parameters

model:

MastraModelConfig

The LLM model to use for evaluating tool appropriateness

availableTools:

Tool[]

Tools available to the agent, for example created with createTool(). Each tool's id and description are passed to the judge as context.

Features

The LLM-based scorer provides:

• Semantic Evaluation: Understands context and user intent
• Appropriateness Assessment: Distinguishes between "helpful" and "appropriate" tools
• Clarification Handling: Recognizes when agents appropriately ask for clarification
• Missing Tool Detection: Identifies tools that should have been called
• Reasoning Generation: Provides explanations for scoring decisions

Evaluation Process

1. Extract Tool Calls: Identifies tools mentioned in agent output
2. Analyze Appropriateness: Evaluates each tool against user request
3. Generate Score: Calculates score based on appropriate vs total tool calls
4. Generate Reasoning: Provides human-readable explanation

LLM-based scoring details

• Fractional scores: Returns values between 0.0 and 1.0
• Context-aware: Considers user intent and appropriateness
• Explanatory: Provides reasoning for scores

LLM-Based Scorer Options

// Basic configuration
const basicLLMScorer = createLLMScorer({
  model: 'openai/gpt-5.6-sol',
  availableTools: [
    createTool({ id: 'tool1', description: 'Description 1' }),
    createTool({ id: 'tool2', description: 'Description 2' })
  ]
});

// With different model
const customModelScorer = createLLMScorer({
  model: 'openai/gpt-5', // More powerful model for complex evaluations
  availableTools: [...]
});

LLM-Based Scorer Results

{
  runId: string,
  score: number,  // 0.0 to 1.0
  reason: string, // Human-readable explanation
  analyzeStepResult: {
    evaluations: Array<{
      toolCalled: string,
      wasAppropriate: boolean,
      reasoning: string
    }>,
    missingTools?: string[]
  }
}

LLM-based scorer examples

The LLM-based scorer uses AI to evaluate whether tool selections are appropriate for the user's request.

Basic LLM evaluation

src/example-llm-basic.ts

import { createTool } from '@mastra/core/tools'
import { createToolCallAccuracyScorerLLM } from '@mastra/evals/scorers/prebuilt'

const llmScorer = createToolCallAccuracyScorerLLM({
  model: 'openai/gpt-5.6-sol',
  availableTools: [
    createTool({
      id: 'weather-tool',
      description: 'Get current weather information for any location',
    }),
    createTool({
      id: 'calendar-tool',
      description: 'Check calendar events and scheduling',
    }),
    createTool({
      id: 'search-tool',
      description: 'Search the web for general information',
    }),
  ],
})

const inputMessages = [
  createTestMessage({
    content: 'What is the weather like in San Francisco today?',
    role: 'user',
    id: 'input-1',
  }),
]

const output = [
  createTestMessage({
    content: 'Let me check the current weather for you.',
    role: 'assistant',
    id: 'output-1',
    toolInvocations: [
      createToolInvocation({
        toolCallId: 'call-123',
        toolName: 'weather-tool',
        args: { location: 'San Francisco', date: 'today' },
        result: { temperature: '68°F', condition: 'foggy' },
        state: 'result',
      }),
    ],
  }),
]

const run = createAgentTestRun({ inputMessages, output })
const result = await llmScorer.run(run)

console.log(result.score) // 1.0 - appropriate tool usage
console.log(result.reason) // "The agent correctly used the weather-tool to address the user's request for weather information."

Handling inappropriate tool usage

src/example-llm-inappropriate.ts

const inputMessages = [
  createTestMessage({
    content: 'What is the weather in Tokyo?',
    role: 'user',
    id: 'input-1',
  }),
]

const inappropriateOutput = [
  createTestMessage({
    content: 'Let me search for that information.',
    role: 'assistant',
    id: 'output-1',
    toolInvocations: [
      createToolInvocation({
        toolCallId: 'call-456',
        toolName: 'search-tool', // Less appropriate than weather-tool
        args: { query: 'Tokyo weather' },
        result: { results: ['Tokyo weather data...'] },
        state: 'result',
      }),
    ],
  }),
]

const run = createAgentTestRun({ inputMessages, output: inappropriateOutput })
const result = await llmScorer.run(run)

console.log(result.score) // 0.5 - partially appropriate
console.log(result.reason) // "The agent used search-tool when weather-tool would have been more appropriate for a direct weather query."

Evaluating clarification requests

The LLM scorer recognizes when agents appropriately ask for clarification:

src/example-llm-clarification.ts

const vagueInput = [
  createTestMessage({
    content: 'I need help with something',
    role: 'user',
    id: 'input-1'
  })
];

const clarificationOutput = [
  createTestMessage({
    content: 'I'd be happy to help! Could you please provide more details about what you need assistance with?',
    role: 'assistant',
    id: 'output-1',
    // No tools called - asking for clarification instead
  })
];

const run = createAgentTestRun({
  inputMessages: vagueInput,
  output: clarificationOutput
});
const result = await llmScorer.run(run);

console.log(result.score); // 1.0 - appropriate to ask for clarification
console.log(result.reason); // "The agent appropriately asked for clarification rather than calling tools with insufficient information."

Comparing both scorers

Here's an example using both scorers on the same data:

src/example-comparison.ts

import { createTool } from '@mastra/core/tools'
import {
  createToolCallAccuracyScorerCode as createCodeScorer,
  createToolCallAccuracyScorerLLM as createLLMScorer,
} from '@mastra/evals/scorers/prebuilt'

// Setup both scorers
const codeScorer = createCodeScorer({
  expectedTool: 'weather-tool',
  strictMode: false,
})

const llmScorer = createLLMScorer({
  model: 'openai/gpt-5.6-sol',
  availableTools: [
    createTool({ id: 'weather-tool', description: 'Get weather information' }),
    createTool({ id: 'search-tool', description: 'Search the web' }),
  ],
})

// Test data
const run = createAgentTestRun({
  inputMessages: [
    createTestMessage({
      content: 'What is the weather?',
      role: 'user',
      id: 'input-1',
    }),
  ],
  output: [
    createTestMessage({
      content: 'Let me find that information.',
      role: 'assistant',
      id: 'output-1',
      toolInvocations: [
        createToolInvocation({
          toolCallId: 'call-1',
          toolName: 'search-tool',
          args: { query: 'weather' },
          result: { results: ['weather data'] },
          state: 'result',
        }),
      ],
    }),
  ],
})

// Run both scorers
const codeResult = await codeScorer.run(run)
const llmResult = await llmScorer.run(run)

console.log('Code Scorer:', codeResult.score) // 0 - wrong tool
console.log('LLM Scorer:', llmResult.score) // 0.3 - partially appropriate
console.log('LLM Reason:', llmResult.reason) // Explains why search-tool is less appropriate

Related

• Answer Relevancy Scorer
• Completeness Scorer
• Faithfulness Scorer
• Custom Scorers