Update Evaluation Assertion

PUT

sdk

evaluation-assertions

{id}

import PromptFoundry from '@prompt-foundry/typescript-sdk';

const client = new PromptFoundry({
  apiKey: process.env['PROMPT_FOUNDRY_API_KEY'], // This is the default and can be omitted
});

async function main() {
  const evaluationAssertion = await client.evaluationAssertions.update('1212121', {
    evaluationId: 'evaluationId',
    targetValue: 'targetValue',
    type: 'EXACT_MATCH',
  });

  console.log(evaluationAssertion);
}

main();

{
  "type": "EXACT_MATCH",
  "evaluationId": "<string>",
  "weight": 1,
  "targetValue": "<string>",
  "jsonPath": "$",
  "ignoreCase": false,
  "negate": false,
  "id": "<string>"
}

Path Parameters

string

required

Example:

"1212121"

Body

application/json

type

enum<string>

required

Available options:

EXACT_MATCH

evaluationId

string

required

targetValue

string

required

The value to match.

weight

number

default:1

How heavily to weigh the assertion within the evaluation.

Required range: 0 <= x <= 1

jsonPath

string | null

default:$

A JSON path to use when matching a JSON response.

ignoreCase

boolean

default:false

Whether to ignore case when comparing strings.

negate

boolean

default:false

Whether to negate the assertion. "true" means the assertion must NOT be true.

type

enum<string>

required

Available options:

EXACT_MATCH

evaluationId

string

required

targetValue

string

required

The value to match.

weight

number

default:1

How heavily to weigh the assertion within the evaluation.

Required range: 0 <= x <= 1

jsonPath

string | null

default:$

A JSON path to use when matching a JSON response.

ignoreCase

boolean

default:false

Whether to ignore case when comparing strings.

negate

boolean

default:false

Whether to negate the assertion. "true" means the assertion must NOT be true.

type

enum<string>

required

Available options:

CONTAINS_ALL

evaluationId

string

required

targetValues

string[]

required

List of values any of which may be present.

weight

number

default:1

How heavily to weigh the assertion within the evaluation.

Required range: 0 <= x <= 1

jsonPath

string | null

default:$

A JSON path to use when matching a JSON response.

ignoreCase

boolean

default:false

Whether to ignore case when comparing strings.

negate

boolean

default:false

Whether to negate the assertion. "true" means the assertion must NOT be true.

type

enum<string>

required

Available options:

CONTAINS_ANY

evaluationId

string

required

targetValues

string[]

required

List of values any of which may be present.

weight

number

default:1

How heavily to weigh the assertion within the evaluation.

Required range: 0 <= x <= 1

jsonPath

string | null

default:$

A JSON path to use when matching a JSON response.

ignoreCase

boolean

default:false

Whether to ignore case when comparing strings.

negate

boolean

default:false

Whether to negate the assertion. "true" means the assertion must NOT be true.

type

enum<string>

required

Available options:

STARTS_WITH

evaluationId

string

required

targetValue

string

required

The value that the response should start with.

weight

number

default:1

How heavily to weigh the assertion within the evaluation.

Required range: 0 <= x <= 1

jsonPath

string | null

default:$

A JSON path to use when matching a JSON response.

ignoreCase

boolean

default:false

Whether to ignore case when comparing strings.

negate

boolean

default:false

Whether to negate the assertion. "true" means the assertion must NOT be true.

type

enum<string>

required

Available options:

COST

evaluationId

string

required

targetThreshold

number

required

The cost threshold to be evaluated against.

weight

number

default:1

How heavily to weigh the assertion within the evaluation.

Required range: 0 <= x <= 1

type

enum<string>

required

Available options:

LATENCY

evaluationId

string

required

targetThreshold

number

required

The latency threshold to be evaluated against.

weight

number

default:1

How heavily to weigh the assertion within the evaluation.

Required range: 0 <= x <= 1

type

enum<string>

required

Available options:

TOOL_CALLED

evaluationId

string

required

toolName

string

required

The name of the tool that should have been called.

weight

number

default:1

How heavily to weigh the assertion within the evaluation.

Required range: 0 <= x <= 1

type

enum<string>

required

Available options:

TOOL_CALLED_WITH

evaluationId

string

required

argKeyName

string

required

The argument name to be matched.

toolName

string

required

The name of the tool that should have been called.

weight

number

default:1

How heavily to weigh the assertion within the evaluation.

Required range: 0 <= x <= 1

ignoreCase

boolean

default:false

Whether to ignore case when comparing argument names.

negate

boolean

default:false

Whether to negate the assertion. "true" means the assertion must NOT be true.

Response

200

application/json

Successful operation

type

enum<string>

required

Available options:

EXACT_MATCH

evaluationId

string

required

targetValue

string

required

The value to match.

string

required

weight

number

default:1

How heavily to weigh the assertion within the evaluation.

Required range: 0 <= x <= 1

jsonPath

string | null

default:$

A JSON path to use when matching a JSON response.

ignoreCase

boolean

default:false

Whether to ignore case when comparing strings.

negate

boolean

default:false

Whether to negate the assertion. "true" means the assertion must NOT be true.

type

enum<string>

required

Available options:

EXACT_MATCH

evaluationId

string

required

targetValue

string

required

The value to match.

string

required

weight

number

default:1

How heavily to weigh the assertion within the evaluation.

Required range: 0 <= x <= 1

jsonPath

string | null

default:$

A JSON path to use when matching a JSON response.

ignoreCase

boolean

default:false

Whether to ignore case when comparing strings.

negate

boolean

default:false

Whether to negate the assertion. "true" means the assertion must NOT be true.

type

enum<string>

required

Available options:

CONTAINS_ALL

evaluationId

string

required

targetValues

string[]

required

List of values any of which may be present.

string

required

weight

number

default:1

How heavily to weigh the assertion within the evaluation.

Required range: 0 <= x <= 1

jsonPath

string | null

default:$

A JSON path to use when matching a JSON response.

ignoreCase

boolean

default:false

Whether to ignore case when comparing strings.

negate

boolean

default:false

Whether to negate the assertion. "true" means the assertion must NOT be true.

type

enum<string>

required

Available options:

CONTAINS_ANY

evaluationId

string

required

targetValues

string[]

required

List of values any of which may be present.

string

required

weight

number

default:1

How heavily to weigh the assertion within the evaluation.

Required range: 0 <= x <= 1

jsonPath

string | null

default:$

A JSON path to use when matching a JSON response.

ignoreCase

boolean

default:false

Whether to ignore case when comparing strings.

negate

boolean

default:false

Whether to negate the assertion. "true" means the assertion must NOT be true.

type

enum<string>

required

Available options:

STARTS_WITH

evaluationId

string

required

targetValue

string

required

The value that the response should start with.

string

required

weight

number

default:1

How heavily to weigh the assertion within the evaluation.

Required range: 0 <= x <= 1

jsonPath

string | null

default:$

A JSON path to use when matching a JSON response.

ignoreCase

boolean

default:false

Whether to ignore case when comparing strings.

negate

boolean

default:false

Whether to negate the assertion. "true" means the assertion must NOT be true.

type

enum<string>

required

Available options:

COST

evaluationId

string

required

targetThreshold

number

required

The cost threshold to be evaluated against.

string

required

weight

number

default:1

How heavily to weigh the assertion within the evaluation.

Required range: 0 <= x <= 1

type

enum<string>

required

Available options:

LATENCY

evaluationId

string

required

targetThreshold

number

required

The latency threshold to be evaluated against.

string

required

weight

number

default:1

How heavily to weigh the assertion within the evaluation.

Required range: 0 <= x <= 1

type

enum<string>

required

Available options:

TOOL_CALLED

evaluationId

string

required

toolName

string

required

The name of the tool that should have been called.

string

required

weight

number

default:1

How heavily to weigh the assertion within the evaluation.

Required range: 0 <= x <= 1

type

enum<string>

required

Available options:

TOOL_CALLED_WITH

evaluationId

string

required

argKeyName

string

required

The argument name to be matched.

toolName

string

required

The name of the tool that should have been called.

string

required

weight

number

default:1

How heavily to weigh the assertion within the evaluation.

Required range: 0 <= x <= 1

ignoreCase

boolean

default:false

Whether to ignore case when comparing argument names.

negate

boolean

default:false

Whether to negate the assertion. "true" means the assertion must NOT be true.

Was this page helpful?

Suggest edits Raise issue

Get Evaluation Assertion Delete Evaluation Assertion

import PromptFoundry from '@prompt-foundry/typescript-sdk';

const client = new PromptFoundry({
  apiKey: process.env['PROMPT_FOUNDRY_API_KEY'], // This is the default and can be omitted
});

async function main() {
  const evaluationAssertion = await client.evaluationAssertions.update('1212121', {
    evaluationId: 'evaluationId',
    targetValue: 'targetValue',
    type: 'EXACT_MATCH',
  });

  console.log(evaluationAssertion);
}

main();

{
  "type": "EXACT_MATCH",
  "evaluationId": "<string>",
  "weight": 1,
  "targetValue": "<string>",
  "jsonPath": "$",
  "ignoreCase": false,
  "negate": false,
  "id": "<string>"
}

API Reference

Path Parameters

Body

Response