flex-md

Flex-MD (v4.0) — Markdown Output Contract with Smart Token Estimation

Flex-MD is a TypeScript library for building and enforcing Markdown Output Contracts with LLMs. It treats Markdown as a semi-structured data format, allowing you to define required sections, list types, and tables while maintaining 100% standard Markdown compatibility.

What's New in v4.0:

• Automatic Token Estimation: Calculate max_tokens directly from your spec
• System Parts Protocol: Standardized size hints that guide LLMs AND enable token prediction
• Smart Toolbox: Cognitive cost analysis, confidence scoring, and improvement detection
• Auto-Fix: Automatically improve specs with one command
• Markdown to JSON: Convert sectioned MD to structured objects with camel-casing (v4.1)

Key Features

Core (v3.0)

• Standard Markdown: No proprietary tags. Pure headings, lists, and tables.
• Strictness Levels (L0–L3): From loose guidance to rigid structural enforcement.
• Deterministic Repair: Auto-fixes misformatted LLM output (merged fences, missing headings, format conversion).
• Instructions Output Format Guidance: Generate formal "Instructions Blocks" for LLM prompts directly from spec objects.
• Issues Envelope: A structured failure format for when repairs fail, allowing safe fallbacks.

Smart Features (v4.0)

• Token Estimation: Automatically calculate max_tokens for API calls based on your spec
• System Parts: Structured instruction patterns (Length: 2-3 paragraphs, Items: 3-5) that guide LLMs and enable estimation
• Compliance Checking: Validate specs meet quality standards (L0-L3 compliance levels)
• Cognitive Cost Analysis: Measure how much effort your spec requires to write/maintain
• Confidence Scoring: Know how accurate your token estimates will be
• Improvement Detection: Find issues and get actionable suggestions
• Auto-Fix: Apply improvements automatically
• Markdown to JSON: Transform sectioned Markdown into structured JSON with camel-cased keys

Installation

npm install flex-md

Runtime Compatibility

• Node LTS baseline: >=18 (see engines in package.json).
• Dual-surface consumption guidance for library authors is documented in docs/consumption.md.

Quick Start

1. Define your Output Format Spec (OFS) with System Parts

import { parseOutputFormatSpec, getMaxTokens } from 'flex-md';

const spec = parseOutputFormatSpec(`
## Output format
- Short answer — text (required)
  Length: 1-2 sentences. Be concise and direct.

- Reasoning — ordered list (required)
  Items: 3-5. Explain your logic step by step.

- Assumptions — list (optional)
  Items: at least 2. List any key assumptions made.

empty sections:
- If a section is empty, write \`None\`.
`);

// Automatically estimate max_tokens needed
const maxTokens = getMaxTokens(spec);
console.log(`Estimated max_tokens: ${maxTokens}`); // ~650

2. Use in Your LLM API Call

const response = await fetch('https://api.anthropic.com/v1/messages', {
  method: 'POST',
  headers: { 
    'Content-Type': 'application/json',
    'x-api-key': API_KEY 
  },
  body: JSON.stringify({
    model: 'claude-sonnet-4-20250514',
    max_tokens: maxTokens,  // Automatically calculated!
    messages: [{
      role: 'user',
      content: yourPrompt + '\n\n' + buildMarkdownGuidance(spec)
    }]
  })
});

3. Enforce the Contract

import { enforceFlexMd } from 'flex-md';

const llmResponse = await response.json();
const result = enforceFlexMd(llmResponse.content[0].text, spec, { level: 2 });

if (result.ok) {
  console.log(result.extracted.sectionsByName["Short answer"].md);
  console.log(result.extracted.sectionsByName["Reasoning"].md);
} else {
  console.log(result.outputText); // Issues Envelope
}

System Parts Protocol

System Parts are structured prefixes in section instructions that serve dual purposes:

1. Guide the LLM on expected output size
2. Enable token estimation for max_tokens calculation

Syntax

[SYSTEM_PART]. [OPTIONAL_GUIDANCE]

Examples:

// Text sections
"Length: 2-3 paragraphs. Provide detailed analysis."
"Length: brief. Keep it short."

// Lists
"Items: 3-5. Focus on key insights."
"Items: at least 3. Be comprehensive."

// Tables
"Rows: 5-7, Columns: 3. Include metrics."

// Code
"Lines: 20-30. Include error handling."
"Lines: ~50. Provide complete example."

Allowed Values

Section Type	System Part Pattern	Examples
text	Length: <value>	brief, moderate, detailed, extensive, 1-2 sentences, 2-3 paragraphs
list	Items: <value>	3, 3-5, at least 3
table	Rows: <value>, Columns: <value>	Rows: 5, Columns: 3, Rows: 3-5, Columns: 4
code	Lines: <value>	20, 15-25, ~50

See System Parts Guide for complete reference.

Compliance Levels (for Spec Authors)

Compliance levels measure how much detail you provide in system parts:

Level	Detail	Cognitive Load	Token Estimation Accuracy
L0	No system parts	None	Fallback (~±40%)
L1	Simple values	Minimal	Basic (~±30%)
L2	Ranges allowed	Low	Good (~±20%)
L3	Full spec with "at least", "~"	Medium	Precise (~±10%)

L2 is recommended for most use cases - good balance of effort and accuracy.

Examples by Level

// L0 - No system parts (fallback estimation)
"Just provide a summary."

// L1 - Simple values
"Length: brief. Provide a summary."
"Items: 3. List the main points."

// L2 - Ranges
"Length: 2-3 paragraphs. Provide detailed analysis."
"Items: 3-5. List key insights."

// L3 - Full specification
"Items: at least 5. Include all relevant factors."
"Lines: ~50. Provide a complete working example."

Input vs Output Format Specs

Flex-MD supports both Input Format Specs and Output Format Specs, but they serve different purposes:

Input Format Specs (Planning & Design)

Input Format Specs (defined with ## Input format) are design-time tools for planning and documentation:

• Documentation: Clearly specify what input structure your LLM expects
• Planning: Help design your prompt structure and data flow
• Coding: Guide developers on how to structure inputs
• Validation: Can be used to validate input data (optional)

Input format specs are NOT used for runtime token estimation — they're for human understanding and system design.

Output Format Specs (Runtime)

Output Format Specs (defined with ## Output format) are runtime tools:

• Token Estimation: Used to calculate max_tokens for API calls
• Validation: Enforce structure on LLM responses
• Extraction: Parse structured data from responses
• Guidance: Tell the LLM exactly what format to return

Output format specs ARE used for runtime token estimation — they directly impact API call parameters.

Example: Using Both

const instructions = `
## Input format
- User Query — text (required)
  The question or request from the user.
- Context — list (optional)
  Items: 1-5. Additional context items.

## Output format
- Answer — text (required)
  Length: 2-3 paragraphs. Provide a comprehensive answer.
- Sources — list (required)
  Items: 3-7. List information sources used.
`;

// Parse both formats
import { parseFormatSpecs } from 'flex-md';
const { input, output } = parseFormatSpecs(instructions);

// Input format: Use for documentation/planning
console.log('Expected input structure:', input);

// Output format: Use for runtime token estimation
const maxTokens = getMaxTokens(output);

Smart Toolbox

Token Estimation

Planning-Time Estimation

For estimating tokens from format specs directly (useful during development):

import { getMaxTokens, estimateSpecTokens } from 'flex-md';

// Quick estimate from output format spec
const maxTokens = getMaxTokens(spec);

// Detailed estimate with options
const estimate = estimateSpecTokens(spec, {
  includeOptional: true,
  safetyMultiplier: 1.3,
  strategy: 'average'  // 'conservative' | 'average' | 'generous'
});

console.log(estimate);
// {
//   total: { estimated: 650, min: 520, max: 780, confidence: 'high' },
//   bySectionName: { ... },
//   overhead: 60
// }

Runtime Estimation

For estimating tokens at runtime with actual prompt, context, and instructions:

import { runtimeEstimateTokens } from 'flex-md';

const estimate = runtimeEstimateTokens({
  prompt: 'Analyze this code and explain what it does.',
  context: 'Previous conversation about TypeScript...',
  instructions: `
    You are a helpful coding assistant.
    
    ## Output format
    - Analysis — text (required)
      Length: 2-3 paragraphs. Explain the code functionality.
    - Code Quality — list (required)
      Items: 3-5. List quality observations.
  `,
  options: {
    safetyMultiplier: 1.2,
    strategy: 'average',
    additionalOverhead: 50  // For system messages, formatting, etc.
  }
});

console.log(estimate.maxTokens);  // Recommended max_tokens for API call (output tokens only)
console.log(estimate.breakdown);
// {
//   prompt: 12,                    // Input tokens (for budgeting)
//   context: 8,                    // Input tokens (for budgeting)
//   instructions: 45,              // Input tokens (for budgeting)
//   output: { total: { estimated: 450, ... }, ... },  // Output token estimate
//   additionalOverhead: 50,
//   total: 565                     // Total tokens (input + output) for budgeting
// }

// Use in API call
const response = await fetch('https://api.anthropic.com/v1/messages', {
  method: 'POST',
  body: JSON.stringify({
    model: 'claude-sonnet-4-20250514',
    max_tokens: estimate.maxTokens,  // Output tokens only (extracted from output format spec)
    messages: [{
      role: 'user',
      content: prompt + '\n\n' + instructions
    }]
  })
});

The runtimeEstimateTokens function:

• Extracts output format specs from instructions automatically (input format specs are ignored)
• Estimates tokens for prompt, context, and instructions text (for budgeting/planning)
• Estimates output tokens based on the output format spec (this becomes max_tokens)
• Provides a breakdown showing input tokens (for budgeting) and output tokens (for API parameter)
• Returns maxTokens which is the value to use for the max_tokens API parameter (output only)

Compliance Checking

import { checkCompliance, formatComplianceReport } from 'flex-md';

const report = checkCompliance(spec, 2); // Check if meets L2

console.log(formatComplianceReport(report));
// Shows which sections need improvement to meet target level

Confidence Scoring

import { calculateConfidence } from 'flex-md';

const confidence = calculateConfidence(spec);

console.log(`Confidence: ${confidence.grade} (${confidence.overall}%)`);
console.log('Recommendations:', confidence.recommendations);
// Grade: B (82%)
// Recommendations: ["Good confidence, but can be improved", ...]

Cognitive Cost Analysis

import { calculateCognitiveCost } from 'flex-md';

const cost = calculateCognitiveCost(spec);

console.log(`Cost: ${cost.totalCost}/100`);
console.log(`Assessment: ${cost.recommendation}`);
// Cost: 28/100
// Assessment: "Moderate cognitive load - reasonable effort required"

Improvement Detection

import { detectImprovements, formatImprovementReport, autoFix } from 'flex-md';

// Detect issues and opportunities
const analysis = detectImprovements(spec, 2);
console.log(formatImprovementReport(analysis));

// Auto-fix quick wins
const fixResult = autoFix(spec, analysis.improvements, {
  applyQuickWinsOnly: true
});

console.log(fixResult.summary);
// "Applied 4 fixes, skipped 1"

Complete Smart Analysis

import { analyzeSpec, formatSmartReport } from 'flex-md';

const analysis = analyzeSpec(spec, 2);
console.log(formatSmartReport(analysis));

Output:

╔═══════════════════════════════════════════════════╗
║         FLEX-MD SMART ANALYSIS REPORT            ║
╚═══════════════════════════════════════════════════╝

📊 SUMMARY DASHBOARD
──────────────────────────────────────────────────
Compliance:      ✓ L2 PASS
Confidence:      B (82%)
Cognitive Cost:  28/100
Token Estimate:  650 tokens

💡 RECOMMENDATIONS
──────────────────────────────────────────────────
🟢 Low Priority:
   • Good confidence, but can be improved
     → Upgrade a few sections to L3 for better precision
...

Strictness Levels (for LLM Output Enforcement)

Level	Goal	Guidance	Enforcement
L0	Plain Markdown	"Reply in Markdown."	None. Accept as-is.
L1	Sectioned MD	"Include these headings..."	Headings must exist.
L2	Fenced Container	"Return inside a single block..."	Exactly one fenced block.
L3	Typed Structure	"Reasoning is an ordered list..."	Enforce list/table kinds.

Note: These are different from Compliance Levels (which measure spec quality) - Strictness Levels control how strictly Flex-MD enforces the contract on LLM output.

The Repair Pipeline

Flex-MD doesn't just validate; it repairs. Our deterministic 9-step plan handles:

1. Container Normalization: Wrapping or merging multiple fenced blocks.
2. Heading Standardization: Case-insensitive matching and naming cleanup.
3. Missing Headings: Adding required sections as None.
4. Stray Content: Moving text outside headings into a default section.
5. Format Conversion: Transforming bullets to numbered lists (and vice-versa) based on spec.

Real-World Example

import { 
  parseOutputFormatSpec, 
  getMaxTokens, 
  analyzeSpec,
  enforceFlexMd 
} from 'flex-md';

// 1. Define spec with system parts
const spec = parseOutputFormatSpec(`
## Output format
- Executive Summary — text (required)
  Length: 2-3 paragraphs. Summarize findings and recommendations.

- Key Metrics — table (required)
  Rows: 5-7, Columns: 3. Include: Metric, Current, Target.

- Action Items — ordered list (required)
  Items: 5-10. Prioritize by impact.

- Technical Details — code (optional)
  Lines: 20-30. Include implementation examples.
`);

// 2. Analyze spec quality
const analysis = analyzeSpec(spec, 2);
console.log(`Confidence: ${analysis.confidence.grade}`);
console.log(`Max tokens: ${analysis.tokenEstimate.total.estimated}`);

// 3. Use in API call
const response = await anthropic.messages.create({
  model: 'claude-sonnet-4-20250514',
  max_tokens: getMaxTokens(spec, { safetyMultiplier: 1.3 }),
  messages: [{
    role: 'user',
    content: `Analyze Q4 performance.\n\n${buildMarkdownGuidance(spec)}`
  }]
});

// 4. Enforce and extract
const result = enforceFlexMd(response.content[0].text, spec, { level: 2 });

if (result.ok) {
  const summary = result.extracted.sectionsByName["Executive Summary"].md;
  const metrics = result.extracted.sectionsByName["Key Metrics"].md;
  // Use structured output...
}

Advanced Usage

Custom Token Estimation

const estimate = estimateSpecTokens(spec, {
  includeOptional: false,      // Skip optional sections
  safetyMultiplier: 1.5,       // Extra headroom
  strategy: 'conservative'     // Use minimum estimates
});

CI/CD Integration

// validate-specs.ts
import { analyzeSpec } from 'flex-md';

const analysis = analyzeSpec(spec, 2);

const highPriorityIssues = analysis.recommendations
  .filter(r => r.priority === 'high');

if (highPriorityIssues.length > 0) {
  console.error('High priority issues found');
  process.exit(1);
}

Progressive Enhancement

Start simple and upgrade as needed:

// Version 1: No system parts (works, but fallback estimation)
const v1 = `
## Output format
- Summary — text (required)
  Write a summary.
`;

// Version 2: Add L1 system parts (better)
const v2 = `
## Output format
- Summary — text (required)
  Length: brief. Write a summary.
`;

// Version 3: Upgrade to L2 (best balance)
const v3 = `
## Output format
- Summary — text (required)
  Length: 2-3 sentences. Write a summary.
`;

API Reference

Core Functions

• parseOutputFormatSpec(markdown) - Parse output format spec from markdown
• parseInputFormatSpec(markdown) - Parse input format spec from markdown
• parseFormatSpecs(instructions) - Extract both input and output format specs from instructions
• stringifyOutputFormatSpec(spec) - Convert spec to markdown
• buildMarkdownGuidance(spec, options) - Generate LLM instructions
• enforceFlexMd(text, spec, options) - Validate and repair LLM output

Token Estimation (v4.0)

• getMaxTokens(spec, options?) - Get estimated max_tokens from spec
• getMaxTokensFromInstructions(instructions, options?) - Extract format specs and estimate tokens
• estimateSpecTokens(spec, options?) - Detailed token estimate from spec
• runtimeEstimateTokens(params) - Runtime estimation with prompt, context, and instructions
• parseSystemPart(instruction, kind) - Parse system part from instruction
• estimateTokens(systemPart) - Estimate tokens for system part

Smart Toolbox (v4.0)

• checkCompliance(spec, level) - Validate compliance level
• calculateConfidence(spec) - Score estimation confidence
• calculateCognitiveCost(spec) - Measure spec complexity
• detectImprovements(spec, level?) - Find issues and suggestions
• autoFix(spec, improvements, options?) - Apply automatic fixes
• analyzeSpec(spec, level?) - Complete smart analysis

Reporting (v4.0)

• formatComplianceReport(report) - Format compliance check
• formatImprovementReport(analysis) - Format improvements
• formatSmartReport(analysis) - Format complete analysis

Markdown to JSON Transformation (v4.1)

Flex-MD includes a robust utility to convert sectioned Markdown (headings and their respective bodies) into a standard JavaScript object/JSON. This is particularly useful for extracting structured data from LLM responses without needing a complex schema.

Features:

• Automatic Camel-Casing: Headings like ### Short Answer or ===Next Steps are automatically converted to valid camelCase keys (shortAnswer, nextSteps).
• Robust Newline Handling: Automatically handles both actual newlines and literal \\n escape sequences often found in LLM outputs.
• Support for All Heading Types: Works with standard ### headings and ===key alternative delimiters.

Usage:

import { markdownToJson } from 'flex-md';

const md = `
### Short Answer
The asset is a server named server1 with private IP 192.168.1.1.

### Next Steps
1. Document in CMDB
2. Perform security check
`;

const data = markdownToJson(md);

console.log(data.shortAnswer); 
// "The asset is a server named server1 with private IP 192.168.1.1."

console.log(data.nextSteps); 
// "1. Document in CMDB\\n2. Perform security check"

NX Flex-MD: Schema-Driven Extraction (Advanced)

For more complex scenarios where you need to enforce a specific JSON schema, handle fuzzy matching of headings, or apply automatic fixes to the data, Flex-MD exposes the NX Flex-MD toolset (powered by nx-md-parser).

Feature Highlights:

• Schema Validation: Define the exact structure and types (string, number, boolean, array, object) you expect.
• Intelligent Heading Matching: Matches headings even if they aren't an exact match (e.g., "Summary" vs "Exec Summary").
• Automatic Type Conversion: Converts Markdown strings to numbers or booleans based on your schema.
• Auto-Fixing: Automatically corrects common formatting issues to match the schema.

Usage:

import { JSONTransformer, Schema } from 'flex-md';

// 1. Define a schema
const schema = Schema.object({
  status: Schema.string(),
  score: Schema.number(),
  isVerified: Schema.boolean(),
  tags: Schema.array(Schema.string())
});

// 2. Create the transformer
const transformer = new JSONTransformer(schema);

// 3. Transform complex Markdown
const md = `
### Status
Active and operational

### Score
95.5

### Verified
Yes

### Tags
- production
- critical
`;

const { result, status, errors } = transformer.transformMarkdown(md);

if (status === 'validated' || status === 'fixed') {
  console.log(result);
  /*
  {
    status: "Active and operational",
    score: 95.5,
    isVerified: true,
    tags: ["production", "critical"]
  }
  */
}

Flex-MD allows you to use its native Output Format Spec (OFS) as the source of truth for NX-MD-Parser's structured extraction.

Modern "Smart" Transformation

The transformWithOfs function is now smarter. It performs dual parsing:

1. Automatic Parsing: Even if you don't have a spec, it extracts all sections and camel-cases keys.
2. Contract Enforcement: If you provide a spec, it uses nx-md-parser to validate, type-cast (lists/tables), and repair the output.

Usage with LLM Outputs:

When working with LLMs, pass the entire response text directly. Flex-MD handles internal normalization (like escaped \n characters) automatically.

import { transformWithOfs } from 'flex-md';

// Pass the RAW content string from your LLM provider
const { 
  parsedOutput,   // Always populated (auto-extraction)
  contractOutput, // Populated if spec was provided
  contractStatus, // "ok" | "different" | "skipped"
  status          // "validated" | "fixed" | "failed"
} = transformWithOfs(llmResponseText, spec);

console.log(parsedOutput.shortAnswer);

Why use this?

1. Zero-Config Extraction: Get structured data without writing a schema first.
2. Dual-Safe: Compare what the LLM sent (parsedOutput) with what the contract required (contractOutput).
3. Internal Normalization: Handles messy data (escaped newlines, merged code blocks) so you don't have to.
4. Fuzzy Matching: Even if the LLM slightly changes the heading (e.g., "Summary" vs "Executive Summary"), the contract will correctly map it.

Advanced AI Features (via NX-MD-Parser 1.4.0)

Flex-MD utilizes the full power of nx-md-parser v1.4.0, providing enterprise-grade AI transformation capabilities.

Multi-Algorithm Fuzzy Matching

The engine uses a weighted combination of four powerful algorithms to find the best match for your headings and keys:

• Jaro-Winkler: Character-level similarity (40%)
• Jaccard Tokens: Token-based similarity (30%)
• Dice Coefficient: N-gram similarity (20%)
• Levenshtein Ratio: Edit distance (10%)

Machine Learning (Learn Aliases)

You can let the system learn from your data to improve matching over time.

import { learnAliasesFromTransformations } from 'flex-md';

const learningResult = learnAliasesFromTransformations([
  {
    input: { "Projct Name": "Test" },
    output: { title: "Test" },
    schema: yourSchema
  }
]);
// System now knows "Projct Name" is an alias for "title"

Intelligent Auto-Fixing

• Typo Correction: Automatically fixes property name typos.
• Structural Repair: Restructures flat objects into nested schemas.
• Smart Conversion: Automatically handles string -> number, string -> boolean, and wrapper types.

Spec Memory: Remember & Recall

Flex-MD includes an in-memory storage feature that allows you to "remember" an Output Format Spec and later reuse it by a unique recallId. This is especially useful for maintaining state within a single execution environment.

Usage:

import { remember, transformWithOfs } from 'flex-md';

// 1. Remember a spec extracted from instructions
const instructions = `
## Output format
- Confidence — number (required)
- Reason — text (required)
`;
const recallId = remember(instructions); // Returns "unique-uuid-..."

// 2. Later, use the recallId instead of the spec object
const md = `
### Confidence
0.95

### Reason
Everything looks correctly formatted based on initial evidence.
`;

const { contractOutput } = transformWithOfs(md, recallId);
console.log(contractOutput.Confidence); // 0.95

Why use this?

• Cleaner Code: Passes simple strings instead of complex spec objects.
• Reference-based workflow: Useful when multiple parts of your system need to agree on the same specification.
• Efficiency: The spec is parsed once and reused.

Documentation

Detailed guides can be found in the docs folder:

• System Parts Guide - Complete protocol reference
• Token Estimation Guide - How estimation works
• Smart Toolbox Guide - Using analysis features
• MDFlex Compliance Spec - Output enforcement

Output Format Spec (OFS) Syntax Reference

Flex-MD uses a simple Markdown block to define the expected output contract. This block is both machine-readable (for validation) and human-readable (for the LLM).

1. Basic Structure

Start with ## Output format. List sections using - Name — Kind.

## Output format
- Summary — text (required)
- Reasoning — ordered list (required)
- Key Tags — list (optional)

2. Available Kinds

Kind	Description	Validation Rule
text (or prose)	Any text content.	Always matches if section exists.
list	Unordered bullets.	Body must contain - items.
ordered list	Numbered list.	Body must contain 1. items.
table	Markdown table.	Must match column headers.

3. Tables

Define tables by specifying columns in parentheses.

Tables:
- (Name, Age, Role — table)
- (Rank, Team, Score — ordered table)

• table: Standard Markdown pipe table.
• ordered table: Must have a first column # with row numbers 1..N.

4. Instructions & Constraints

You can add instructions under any section item to guide the LLM.

- Summary — text
    Length: 2-3 sentences. No bullet points.

5. Empty Sections

Define what to write if a section has no content:

Empty sections:
- If a section is empty, write `None`.

6. Complete Example

Here is a full real-world example from a security analysis tool:

## Output format
- Executive Summary — text (required)
  Brief overview of findings.
- Critical Vulnerabilities — ordered list (required)
- Remediation Plan — text (required)

Tables:
- (Component, Severity, Fix — table)

Empty sections:
- If a section is empty, write `None`.

Migration from v3.0

v4.0 is 100% backwards compatible with v3.0. All existing code continues to work.

To adopt v4.0 features:

1. Add system parts to your spec instructions:

   - Summary — text (required)
   -   Provide a brief overview.
   + Summary — text (required)
   +   Length: 2-3 sentences. Provide a brief overview.

2. Use token estimation:

   const maxTokens = getMaxTokens(spec);

3. Analyze and improve your specs:

   const analysis = analyzeSpec(spec);
   console.log(formatSmartReport(analysis));

Why Flex-MD v4.0?

Before v4.0

// Guessing max_tokens
const response = await api.create({
  max_tokens: 2000,  // 🤷 Is this enough? Too much?
  ...
});

After v4.0

// Precise estimation
const maxTokens = getMaxTokens(spec);  // ✓ 650 tokens (±20%)
const response = await api.create({
  max_tokens,  // 🎯 Right-sized
  ...
});

Benefits

• Faster responses: Right-sized tokens mean lower latency
• Lower costs: Don't overpay for unused tokens
• Better accuracy: Clear size expectations guide LLMs
• Quality insights: Know your spec's strengths/weaknesses
• Easy maintenance: Auto-detect and fix issues

License

MIT

---

Flex-MD v4.0 - Smart Markdown contracts for production LLM applications.