Speak Core Workflow A

What it does

When your agent uses it

• Implementing conversation practice features
• Building AI tutor interactions
• Core language learning dialogue systems
• With phrases like speak conversation practice

Example prompts

• “speak conversation practice”
• “speak AI tutor”
• “speak dialogue”
• “/speak-core-workflow-a”

Requirements

• A credential in SPEAK_API_KEY
• Compatibility (from SKILL.md): Designed for Claude Code
• Pre-approved tools (allowed-tools): Read, Write, Edit, Bash(npm:*), Grep

Workflow steps

4 steps, taken from the step headings in SKILL.md.

1. 1Start a Conversation Session
2. 2Send Student Responses
3. 3Conversation Loop with Progress Tracking
4. 4Multi-Topic Session

What it can do on your machine

Read from SKILL.md and the folder at commit cfae287. It shows what the files ask for, not the result of running them.

• Tool permissions

  Pre-approves these tools, so the agent can use them without asking each time:

  • Read
  • Write
  • Edit
  • Bash(npm:*)
  • Grep

  From allowed-tools in the SKILL.md frontmatter.

• Runs code

  No scripts in the folder and no shell commands in SKILL.md (its code samples are typescript).

  From the folder's file list and the shell code blocks in SKILL.md.

• Network

  Links to these hosts (documentation or services it may open):

  • speak.com
  • platform.openai.com

  From URLs in SKILL.md, links to its own repository left out.

• Credentials

  Names these keys or tokens, usually read from environment variables:

  • SPEAK_API_KEY

  From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.

• Compatibility

  Designed for Claude Code

  From compatibility in the SKILL.md frontmatter.

Context cost

Speak Core Workflow A loads about 1.5k tokens when it runs, and up to ~3.2k if it reads all its reference files. Until then it costs about 88 tokens; SKILL.md has 240 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it

~88

When it runs · the whole SKILL.md, loaded when a task matches

~1.5k

With references · SKILL.md plus every file in references/, read only if the agent opens them

~3.2k

Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.

Safety

Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); files beside SKILL.md are not scanned.

SKILL.md

The full file from jeremylongshore/tons-of-skills-marketplace at commit cfae287, republished under its MIT licence (© jeremylongshore). 240 words, ~1,526 tokens.

name

speak-core-workflow-a

description

Execute Speak primary workflow: AI Conversation Practice with real-time feedback. Use when implementing conversation practice features, building AI tutor interactions, or core language learning dialogue systems. Trigger with phrases like "speak conversation practice", "speak AI tutor", "speak dialogue", "primary speak workflow".

allowed-tools

Read, Write, Edit, Bash(npm:*), Grep

compatibility

Designed for Claude Code

version

1.14.0

license

MIT

author

Jeremy Longshore <jeremy@intentsolutions.io>

tags

saas, speak, workflow

import { SpeakClient } from '@speak/language-sdk';

const client = new SpeakClient({
  apiKey: process.env.SPEAK_API_KEY!,
  appId: process.env.SPEAK_APP_ID!,
  language: 'es',
});

// Start a restaurant ordering scenario in Spanish
const session = await client.startConversation({
  scenario: 'ordering-food',
  language: 'es',
  level: 'intermediate',
  nativeLanguage: 'en',
  maxTurns: 10,
  feedbackDetail: 'phoneme', // 'word' or 'phoneme'
});

console.log('Session started:', session.id);
console.log('AI Tutor:', session.firstPrompt.text);
// "Bienvenido al restaurante. Soy tu camarero. Que le gustaria ordenar?"

// Submit audio for pronunciation scoring
const turn1 = await client.sendTurn(session.id, {
  audioPath: './recordings/student-response-1.wav',
});

console.log('Tutor:', turn1.tutorText);
console.log('Pronunciation:', turn1.pronunciationScore); // 0-100
console.log('Grammar:', turn1.corrections);
// [{original: "yo quiero", suggestion: "quisiera", note: "More polite form for ordering"}]
console.log('Vocabulary:', turn1.vocabularyNotes);
// ["camarero = waiter", "ordenar = to order"]

// Or submit text (skips pronunciation scoring)
const turn2 = await client.sendTurn(session.id, {
  text: 'Quisiera una ensalada y un vaso de agua, por favor.',
});

async function runConversationLesson(
  client: SpeakClient,
  scenario: string,
  language: string,
  level: string,
) {
  const session = await client.startConversation({
    scenario, language, level, nativeLanguage: 'en',
  });

  const turns: TurnResult[] = [];
  let isComplete = false;

  while (!isComplete && turns.length < 10) {
    // Display tutor prompt
    const prompt = turns.length === 0
      ? session.firstPrompt.text
      : turns[turns.length - 1].tutorText;
    console.log(`\nTutor: ${prompt}`);

    // Get student audio (mic input or file)
    const audioPath = await recordStudentAudio();

    // Submit and get feedback
    const turn = await client.sendTurn(session.id, { audioPath });
    turns.push(turn);

    // Show feedback
    if (turn.pronunciationScore < 60) {
      console.log(`Pronunciation needs work: ${turn.pronunciationScore}/100`);
      console.log('Try again with this phrase.');
    }
    if (turn.corrections.length > 0) {
      console.log('Grammar notes:', turn.corrections.map(c => c.note).join('; '));
    }

    isComplete = turn.sessionComplete;
  }

  // End session and get summary
  const summary = await client.endSession(session.id);
  return summary;
}

const topics = ['greetings', 'directions', 'ordering-food', 'shopping'];
const results: SessionSummary[] = [];

for (const topic of topics) {
  console.log(`\n=== ${topic.toUpperCase()} ===`);
  const summary = await runConversationLesson(client, topic, 'es', 'intermediate');
  results.push(summary);
  console.log(`Score: ${summary.avgPronunciationScore}/100`);
}

// Overall progress report
console.log('\n=== Session Report ===');
console.table(results.map(r => ({
  topic: r.scenario,
  pronunciation: r.avgPronunciationScore,
  grammar: r.grammarAccuracy + '%',
  newWords: r.newWords.length,
  duration: r.durationMinutes + 'min',
})));

© jeremylongshore, MIT. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file

Files

SKILL.md and 1 other file (references) in skills/.curated/speak-core-workflow-a of jeremylongshore/tons-of-skills-marketplace.

• SKILL.md
• references/implementation-guide.md

Open the folder on GitHubat commit cfae287