Recipe catalog / exercise-select
Select the practice exercise that addresses feedback
Which candidate practice exercise in candidates best addresses the problems raised in feedback, given the player's goal?
You are building a practice app that turns a teacher's or an automated assessor's feedback into a concrete assignment chosen from your own exercise library.
Explore this recipe interactively ยท Source and implementation guide
Use exercise-select in TypeScript
Install with npm install jev-recipes. Requires Node.js 22.9 or newer and ES modules. Set TYPESAFE_API_KEY in your server environment for live calls, which send input to TypeSafe and use API quota. See the installation guide.
import { exerciseSelect } from 'jev-recipes/exercise-select';
const result = await exerciseSelect({
"feedback": "You are rushing the sixteenth-note runs in the development section, and the left hand loses its evenness whenever the right hand takes the melody in bars 41 to 56.",
"goal": "Perform the first movement at the studio recital in six weeks.",
"candidates": [
{
"id": "dev-section-metronome",
"text": "Development section (bars 41 to 56) hands separately with the metronome clicking sixteenths at half tempo, then hands together, raising the tempo 4 bpm after each clean pass."
},
{
"id": "scales-in-thirds",
"text": "Major scales in thirds, both hands, four octaves, all keys, at a steady moderate tempo."
}
],
"minConfidence": 0.8
});
console.log(result);
Input contract
| Field | Type | Needed |
|---|---|---|
| feedback | string | Required |
| candidates | array | Required |
| goal | string | Optional |
| minConfidence | number | Optional |
Full input and result schemas
{
"input": {
"$schema": "https://json-schema.org/draft/2020-12/schema",
"type": "object",
"properties": {
"feedback": {
"type": "string"
},
"candidates": {
"minItems": 1,
"maxItems": 50,
"type": "array",
"items": {
"type": "object",
"properties": {
"id": {
"type": "string"
},
"text": {
"type": "string"
}
},
"required": [
"id",
"text"
]
}
},
"goal": {
"type": "string"
},
"minConfidence": {
"type": "number",
"minimum": 0,
"maximum": 1
}
},
"required": [
"feedback",
"candidates"
]
},
"result": {
"$schema": "https://json-schema.org/draft/2020-12/schema",
"type": "object",
"properties": {
"model": {
"type": "string"
},
"usage": {
"type": "object",
"properties": {
"input_tokens": {
"type": "integer",
"minimum": 0,
"maximum": 9007199254740991
},
"output_tokens": {
"type": "integer",
"minimum": 0,
"maximum": 9007199254740991
}
},
"required": [
"input_tokens",
"output_tokens"
],
"additionalProperties": false
},
"status": {
"type": "string",
"enum": [
"ready",
"review"
]
},
"verdict": {
"type": "string",
"enum": [
"matched",
"none",
"ambiguous"
]
},
"selection": {
"type": [
"string",
"null"
]
},
"suggestedSelection": {
"type": [
"string",
"null"
]
},
"confidence": {
"type": "number",
"minimum": 0,
"maximum": 1
},
"probabilities": {
"type": "object",
"properties": {
"candidates": {
"type": "object",
"propertyNames": {
"type": "string"
},
"additionalProperties": {
"type": "number",
"minimum": 0,
"maximum": 1
}
},
"none": {
"type": "number",
"minimum": 0,
"maximum": 1
},
"ambiguous": {
"type": "number",
"minimum": 0,
"maximum": 1
}
},
"required": [
"candidates",
"none",
"ambiguous"
],
"additionalProperties": false
}
},
"required": [
"model",
"usage",
"status",
"verdict",
"selection",
"suggestedSelection",
"confidence",
"probabilities"
],
"additionalProperties": false
}
}Saved example result
This hand-authored response demonstrates the contract. It is not a model accuracy measurement. Run it without an API key: npx jev-recipes demo exercise-select.
{
"model": "demo-fixture",
"usage": {
"input_tokens": 0,
"output_tokens": 0
},
"status": "ready",
"verdict": "matched",
"selection": "dev-section-metronome",
"suggestedSelection": "dev-section-metronome",
"confidence": 0.9,
"probabilities": {
"candidates": {
"dev-section-metronome": 0.9,
"scales-in-thirds": 0.06
},
"none": 0.03,
"ambiguous": 0.01
}
}
Evaluation evidence
No verified live accuracy measurement is available. Evaluate representative cases before using this decision in your workflow.
Use the evaluation guide to measure this decision on your own labeled cases.
Limitations
- Chooses among the supplied candidates only; it does not invent exercises or judge whether the library is adequate.
- Judges fit to the problems as worded in feedback. It does not know the player's history, schedule, or physical limits.
- Supply 1 to 50 candidates, each with a unique non-empty ID and non-empty text.
Related recipes
- choose-action: Use choose-action to pick the next step from a list of general actions, rather than a practice exercise from an exercise library.
- troubleshooting-fit: Use troubleshooting-fit to check whether one supplied procedure addresses a reported problem, rather than to pick the best of several.