GET /find-specs

Search for specifications using powerful regex patterns with optional status and endpoint filtering.

Overview

Cost: Free (no credits required)
Use Cases: Keyword search, pattern matching, finding related specs
Response Time: Instant
Search: Case-insensitive regex matching against spec input text
Returns: Paginated array of matching specs

Endpoint

GET https://api.pre.dev/find-specs

Headers

Authorization: Bearer YOUR_API_KEY

Query Parameters

Parameter	Type	Required	Default	Description
`query`	string	✅	-	REQUIRED - Regex pattern to search (case-insensitive)
`limit`	integer	❌	20	Results per page (1-100)
`skip`	integer	❌	0	Number of records to skip for pagination
`endpoint`	string	❌	-	Filter by endpoint: `fast_spec` or `deep_spec`
`status`	string	❌	-	Filter by status: `pending`, `processing`, `completed`, or `failed`

Parameter Details

query (REQUIRED)

Regex pattern matched against spec input field
Case-insensitive by default
Supports full regex syntax
Must be URL-encoded in request

limit

Minimum: 1
Maximum: 100
Default: 20

skip

Minimum: 0
Used for pagination

endpoint

fast_spec - Only search Fast Spec generations
deep_spec - Only search Deep Spec generations

status

pending - Only queued specs
processing - Only specs currently generating
completed - Only successfully finished specs
failed - Only failed generations

Regex Pattern Examples

Pattern	What It Matches	Example Use Case
`payment`	Contains “payment” (any case)	Find all payment-related specs
`^Build`	Starts with “Build”	Find all “Build X” specs
`platform$`	Ends with “platform”	Find platform projects
`(API\|REST)`	Contains “API” OR “REST”	Find API-related specs
`auth.*system`	”auth” followed by “system”	Find authentication systems
`\d{3,}`	Contains 3+ consecutive digits	Find specs with quantities/budgets
`saas\|sass`	Contains “saas” OR “sass”	Catch common misspellings
`e-?commerce`	”ecommerce” or “e-commerce”	Match hyphen variations
`task.*management`	”task” then “management”	Find task/project mgmt tools
`real.?time`	”realtime” or “real time”	Match spacing variations

Response

Success Response

Same structure as /list-specs:

{
  "specs": [
    {
      "_id": "507f1f77bcf86cd799439011",
      "created": "2024-01-15T14:30:00.000Z",
      "endpoint": "fast_spec",
      "input": "Build a payment processing system with Stripe integration",
      "status": "completed",
      "success": true,
      "humanSpecUrl": "https://api.pre.dev/s/a6hFJRV6",
      "totalHumanHours": 120,
      "codingAgentSpecUrl": "https://api.pre.dev/s/a6hFJRV7",
      "executionTime": 38500
    }
  ],
  "total": 8,
  "hasMore": false
}

Response Fields

Identical to the list-specs endpoint. See that documentation for complete field descriptions.

Code Examples

cURL Examples

Simple keyword search:

curl -H "Authorization: Bearer YOUR_API_KEY" \
  "https://api.pre.dev/find-specs?query=payment"

Search with URL encoding:

# Search for specs starting with "Build"
curl -H "Authorization: Bearer YOUR_API_KEY" \
  "https://api.pre.dev/find-specs?query=%5EBuild"

Search completed specs only:

curl -H "Authorization: Bearer YOUR_API_KEY" \
  "https://api.pre.dev/find-specs?query=dashboard&status=completed"

Search with OR condition:

# Find specs containing "API" OR "REST"
curl -H "Authorization: Bearer YOUR_API_KEY" \
  "https://api.pre.dev/find-specs?query=(API%7CREST)"

Complex pattern with pagination:

# Find e-commerce specs (handles hyphen variations)
curl -H "Authorization: Bearer YOUR_API_KEY" \
  "https://api.pre.dev/find-specs?query=e-?commerce&limit=50&skip=0"

JavaScript/Node.js

async function findSpecs(query, options = {}) {
  const {
    limit = 20,
    skip = 0,
    endpoint = null,
    status = null
  } = options;

  const params = new URLSearchParams({
    query,
    limit: limit.toString(),
    skip: skip.toString()
  });

  if (endpoint) params.append('endpoint', endpoint);
  if (status) params.append('status', status);

  const response = await fetch(
    `https://api.pre.dev/find-specs?${params}`,
    {
      headers: { 'Authorization': 'Bearer YOUR_API_KEY' }
    }
  );

  if (!response.ok) {
    throw new Error(`API error: ${response.status}`);
  }

  return response.json();
}

// Usage examples

// Simple keyword search
const paymentSpecs = await findSpecs('payment');
console.log(`Found ${paymentSpecs.total} payment specs`);

// Search for specs starting with "Build"
const buildSpecs = await findSpecs('^Build');

// Search with filters
const dashboards = await findSpecs('dashboard', {
  status: 'completed',
  limit: 50
});

// Complex regex: find API or REST specs
const apiSpecs = await findSpecs('(API|REST)');

// Search with pagination
async function searchAllMatches(query) {
  let allMatches = [];
  let skip = 0;
  const limit = 50;

  while (true) {
    const response = await findSpecs(query, { skip, limit });
    allMatches.push(...response.specs);
    
    if (!response.hasMore) break;
    skip += limit;
  }

  return allMatches;
}

// Search for e-commerce (with hyphen variation)
const ecommerce = await findSpecs('e-?commerce');

// Find authentication systems
const authSystems = await findSpecs('auth.*system');

TypeScript

interface FindSpecsOptions {
  limit?: number;
  skip?: number;
  endpoint?: 'fast_spec' | 'deep_spec';
  status?: 'pending' | 'processing' | 'completed' | 'failed';
}

interface SearchResult {
  specs: SpecObject[];
  total: number;
  hasMore: boolean;
  query: string;
}

class SpecSearchClient {
  constructor(private apiKey: string) {}

  async findSpecs(
    query: string,
    options: FindSpecsOptions = {}
  ): Promise<SearchResult> {
    const {
      limit = 20,
      skip = 0,
      endpoint,
      status
    } = options;

    const params = new URLSearchParams({
      query,
      limit: limit.toString(),
      skip: skip.toString()
    });

    if (endpoint) params.append('endpoint', endpoint);
    if (status) params.append('status', status);

    const response = await fetch(
      `https://api.pre.dev/find-specs?${params}`,
      {
        headers: { 'Authorization': `Bearer ${this.apiKey}` }
      }
    );

    if (!response.ok) {
      throw new Error(`API error: ${response.status}`);
    }

    const data = await response.json();
    return { ...data, query };
  }

  async *searchPaginated(
    query: string,
    options: FindSpecsOptions = {}
  ): AsyncGenerator<SpecObject[], void, unknown> {
    let skip = 0;
    const limit = options.limit || 50;

    while (true) {
      const response = await this.findSpecs(query, { ...options, skip, limit });
      
      if (response.specs.length === 0) break;
      
      yield response.specs;
      
      if (!response.hasMore) break;
      skip += limit;
    }
  }

  async searchWithHighlight(
    query: string,
    options: FindSpecsOptions = {}
  ): Promise<Array<SpecObject & { highlight: string }>> {
    const result = await this.findSpecs(query, options);
    
    // Add highlighting to matched text
    const regex = new RegExp(`(${query})`, 'gi');
    
    return result.specs.map(spec => ({
      ...spec,
      highlight: spec.input.replace(regex, '<mark>$1</mark>')
    }));
  }
}

// Usage
const searchClient = new SpecSearchClient('YOUR_API_KEY');

// Simple search
const results = await searchClient.findSpecs('payment', { 
  status: 'completed' 
});
console.log(`Found ${results.total} completed payment specs`);

// Paginated search
for await (const batch of searchClient.searchPaginated('dashboard')) {
  console.log(`Processing ${batch.length} dashboard specs...`);
  batch.forEach(spec => console.log(`  - ${spec.input}`));
}

// Search with highlighting
const highlighted = await searchClient.searchWithHighlight('payment');
highlighted.forEach(spec => {
  console.log(spec.highlight); // HTML with <mark> tags
});

// Complex searches
const apiOrRest = await searchClient.findSpecs('(API|REST)');
const startsWithBuild = await searchClient.findSpecs('^Build');
const hasRealtime = await searchClient.findSpecs('real.?time');

Python

import requests
from typing import Optional, List, Dict, Any, Generator
import urllib.parse

class SpecSearchClient:
    def __init__(self, api_key: str):
        self.api_key = api_key
        self.base_url = 'https://api.pre.dev'
    
    def find_specs(
        self,
        query: str,
        limit: int = 20,
        skip: int = 0,
        endpoint: Optional[str] = None,
        status: Optional[str] = None
    ) -> Dict[str, Any]:
        """
        Search specs using regex pattern.
        
        Args:
            query: Regex pattern to search (case-insensitive)
            limit: Number of results per page (1-100)
            skip: Number of records to skip
            endpoint: Filter by 'fast_spec' or 'deep_spec'
            status: Filter by 'pending', 'processing', 'completed', or 'failed'
        
        Returns:
            Dictionary with 'specs' array, 'total' count, and 'hasMore' flag
        """
        params = {
            'query': query,
            'limit': limit,
            'skip': skip
        }
        
        if endpoint:
            params['endpoint'] = endpoint
        if status:
            params['status'] = status
        
        response = requests.get(
            f'{self.base_url}/find-specs',
            params=params,
            headers={'Authorization': f'Bearer {self.api_key}'}
        )
        response.raise_for_status()
        return response.json()
    
    def search_all(
        self,
        query: str,
        endpoint: Optional[str] = None,
        status: Optional[str] = None,
        batch_size: int = 50
    ) -> List[Dict[str, Any]]:
        """Fetch all specs matching the search query."""
        all_results = []
        skip = 0
        
        while True:
            response = self.find_specs(
                query=query,
                limit=batch_size,
                skip=skip,
                endpoint=endpoint,
                status=status
            )
            all_results.extend(response['specs'])
            
            if not response['hasMore']:
                break
            
            skip += batch_size
        
        return all_results
    
    def search_paginated(
        self,
        query: str,
        endpoint: Optional[str] = None,
        status: Optional[str] = None,
        batch_size: int = 50
    ) -> Generator[List[Dict[str, Any]], None, None]:
        """Generator that yields batches of search results."""
        skip = 0
        
        while True:
            response = self.find_specs(
                query=query,
                limit=batch_size,
                skip=skip,
                endpoint=endpoint,
                status=status
            )
            
            if not response['specs']:
                break
            
            yield response['specs']
            
            if not response['hasMore']:
                break
            
            skip += batch_size

# Usage examples
client = SpecSearchClient('YOUR_API_KEY')

# Simple keyword search
payment_specs = client.find_specs('payment')
print(f"Found {payment_specs['total']} payment specs")

# Search with filters
dashboards = client.find_specs(
    'dashboard',
    status='completed',
    limit=50
)

# Complex regex patterns
api_specs = client.find_specs('(API|REST)')  # OR condition
build_specs = client.find_specs('^Build')    # Starts with
auth_systems = client.find_specs('auth.*system')  # Pattern with wildcard

# Search for e-commerce (handles hyphen variations)
ecommerce = client.find_specs('e-?commerce')

# Get all matches
all_payment_specs = client.search_all('payment', status='completed')
print(f"Total payment specs: {len(all_payment_specs)}")

# Process in batches
for batch in client.search_paginated('dashboard', batch_size=25):
    print(f"Processing {len(batch)} dashboard specs")
    for spec in batch:
        print(f"  - {spec['input'][:60]}...")

# Common search patterns
def search_patterns():
    """Examples of useful search patterns."""
    
    # Find all SaaS projects
    saas_specs = client.find_specs('saas', status='completed')
    
    # Find real-time features (handles spacing)
    realtime = client.find_specs('real.?time')
    
    # Find specs with numbers (budgets, quantities)
    with_numbers = client.find_specs(r'\d{3,}')
    
    # Find authentication/authorization
    auth = client.find_specs('auth(entication|orization)?')
    
    # Find e-commerce variations
    ecommerce = client.find_specs('e-?commerce')
    
    return {
        'saas': saas_specs['total'],
        'realtime': realtime['total'],
        'with_numbers': with_numbers['total'],
        'auth': auth['total'],
        'ecommerce': ecommerce['total']
    }

Common Use Cases

1. Search Bar Implementation

async function handleSearch(searchTerm, filters = {}) {
  try {
    // Escape special regex characters for literal search
    const escapedTerm = searchTerm.replace(/[.*+?^${}()|[\]\\]/g, '\\$&');
    
    const results = await findSpecs(escapedTerm, {
      status: filters.status,
      endpoint: filters.endpoint,
      limit: 20
    });
    
    return {
      results: results.specs,
      total: results.total,
      hasMore: results.hasMore
    };
  } catch (error) {
    console.error('Search failed:', error);
    return { results: [], total: 0, hasMore: false };
  }
}

def find_related_specs(base_spec_input: str, client: SpecSearchClient):
    """Find specs related to a given spec."""
    # Extract key terms (simplified example)
    keywords = base_spec_input.lower().split()[:3]
    pattern = '|'.join(keywords)
    
    related = client.find_specs(pattern, limit=10)
    return related['specs']

3. Category-Based Search

const CATEGORY_PATTERNS: Record<string, string> = {
  ecommerce: 'e-?commerce|shop|store|cart',
  authentication: 'auth(entication|orization)?|login|signup',
  realtime: 'real.?time|websocket|live|streaming',
  payment: 'payment|billing|stripe|checkout',
  dashboard: 'dashboard|analytics|metrics|reporting',
  api: '(API|REST|GraphQL)',
  social: 'social|chat|messaging|comment',
  ai: '(AI|ML|machine.?learning|artificial.?intelligence)'
};

async function searchByCategory(category: string) {
  const pattern = CATEGORY_PATTERNS[category];
  if (!pattern) {
    throw new Error(`Unknown category: ${category}`);
  }
  
  return await searchClient.findSpecs(pattern, {
    status: 'completed',
    limit: 50
  });
}

4. Advanced Search with Multiple Criteria

async function advancedSearch({
  mustInclude = [],      // All of these terms
  shouldInclude = [],    // Any of these terms
  mustExclude = [],      // None of these terms
  status = null,
  endpoint = null
}) {
  // Build regex pattern
  let pattern = '';
  
  if (mustInclude.length > 0) {
    // Positive lookahead for each required term
    pattern = mustInclude.map(term => `(?=.*${term})`).join('');
    pattern += '.*';
  }
  
  if (shouldInclude.length > 0) {
    pattern += `(${shouldInclude.join('|')})`;
  }
  
  // Note: mustExclude requires client-side filtering as regex negative lookahead
  // can be complex. Better to filter results after fetching.
  
  const results = await findSpecs(pattern || '.*', { status, endpoint });
  
  // Client-side filtering for exclusions
  if (mustExclude.length > 0) {
    const excludeRegex = new RegExp(mustExclude.join('|'), 'i');
    results.specs = results.specs.filter(
      spec => !excludeRegex.test(spec.input)
    );
    results.total = results.specs.length;
  }
  
  return results;
}

// Usage
const results = await advancedSearch({
  mustInclude: ['task', 'management'],
  shouldInclude: ['team', 'collaboration'],
  mustExclude: ['deprecated'],
  status: 'completed'
});

Regex Tips & Tricks

Common Patterns

// Case variations
'saas|SaaS|SAAS'  // Match any capitalization

// Word boundaries
'\\bapi\\b'       // Match "api" as whole word only

// Optional characters
'e-?commerce'     // Match "ecommerce" or "e-commerce"
'real.?time'      // Match "realtime" or "real time"

// Character classes
'[Pp]ayment'      // Match "Payment" or "payment"
'task[- ]management'  // Match "task management" or "task-management"

// Quantifiers
'\\d{3,}'         // 3 or more digits
'feature.*flag'   // "feature" then "flag" with anything between

// Alternation
'(dashboard|admin|panel)'  // Match any of these

// Start/End anchors
'^Build'          // Must start with "Build"
'platform$'       // Must end with "platform"

URL Encoding

When using cURL or constructing URLs directly, encode special characters:

Character	Encoded	Example Pattern	Encoded URL
`^`	`%5E`	`^Build`	`query=%5EBuild`
`$`	`%24`	`platform$`	`query=platform%24`
`\|`	`%7C`	`API\|REST`	`query=API%7CREST`
`(`	`%28`	`(API\|REST)`	`query=%28API%7CREST%29`
`)`	`%29`	`(API\|REST)`	`query=%28API%7CREST%29`
Space	`%20` or `+`	`task management`	`query=task+management`

Most HTTP clients handle this automatically.

Best Practices

Search Patterns

Start with simple keywords, add complexity if needed
Use case-insensitive patterns (already default)
Handle common variations (hyphens, spaces, plural forms)
Test patterns before using in production

Performance

Use specific patterns to reduce result sets
Combine with status and endpoint filters
Implement pagination for large result sets
Cache frequent searches

User Experience

Show loading states during search
Display result counts
Highlight matched terms in results
Provide search suggestions or examples
Handle empty results gracefully

Error Handling

Validate regex patterns client-side when possible
Catch and display API errors clearly
Provide fallback for invalid regex
Show helpful messages for no results

Limitations

Search only matches against the input field
Maximum 100 results per request (use pagination for more)
Regex is case-insensitive by default
Very complex regex patterns may impact performance

Back: List Specs

List all specs with pagination and filtering.

Authorizations

Authorization

string

header

default:YOUR_API_KEY

required

API key for authentication. Get your API key from https://pre.dev/projects/playground (Solo) or https://pre.dev/enterprise/dashboard?page=api (Enterprise). Use format: Bearer YOUR_API_KEY

Query Parameters

query

string

required

REQUIRED - Regex pattern to search (case-insensitive)

Example:

"payment"

limit

integer

default:20

Number of results per page (1-100)

Required range: 1 <= x <= 100

Example:

20

skip

integer

default:0

Number of records to skip for pagination

Required range: x >= 0

Example:

0

endpoint

enum<string>

Filter by endpoint type

Available options:

fast_spec,

deep_spec

Example:

"fast_spec"

status

enum<string>

Filter by processing status

Available options:

pending,

processing,

completed,

failed

Example:

"completed"

Response

Search results retrieved successfully

specs

object[]

required

Array of spec objects matching the filters

Show child attributes

specs._id

string

MongoDB ObjectId of the spec request

specs.created

string<date-time>

ISO timestamp when the request was created

specs.endpoint

enum<string>

Which endpoint was used

Available options:

fast_spec,

deep_spec

specs.input

string

Original input text provided

specs.status

enum<string>

Current status

Available options:

pending,

processing,

completed,

failed

specs.success

boolean

Whether the request succeeded

specs.uploadedFileShortUrl

string

Short URL for uploaded file (if file was uploaded)

specs.uploadedFileName

string

Name of uploaded file (if file was uploaded)

specs.humanSpecUrl

string<uri>

URL where the human-readable spec is hosted (only when completed)

specs.totalHumanHours

number

Estimated total hours for a human to implement the spec (only when completed)

specs.codingAgentSpecUrl

string<uri>

URL where the coding agent spec format is hosted (only when completed)

specs.executionTime

integer

Processing time in milliseconds (only when completed or failed)

specs.predevUrl

string<uri>

pre.dev project URL (only when completed)

specs.lovableUrl

string<uri>

Deep link to Lovable.dev (only when completed)

specs.cursorUrl

string

Deep link to Cursor (only when completed)

specs.v0Url

string<uri>

Deep link to Vercel v0 (only when completed)

specs.boltUrl

string<uri>

Deep link to Bolt.new (only when completed)

specs.zippedDocsUrls

object[]

Array of scraped documentation archives (only when completed). Empty array if no docURLs provided or scraping fails

Show child attributes

specs.zippedDocsUrls.platform

string

required

Hostname extracted from the documentation URL (e.g., 'stripe.com', 'docs.github.com')

Example:

"stripe.com"

specs.zippedDocsUrls.masterZipShortUrl

string<uri>

required

Short URL to download the zipped documentation archive for this platform

specs.zippedDocsUrls.masterMarkdownShortUrl

string<uri>

Optional short URL to consolidated markdown file for this platform

specs.errorMessage

string

Error description (only when failed)

specs.progress

string

Human-readable progress description

total

integer

required

Total number of specs matching the filters

Example:

42

hasMore

boolean

required

Whether more pages are available

Example:

true

Using the API

Spec Generation

Spec Management

Find Specs

GET /find-specs

Overview

Endpoint

Headers

Query Parameters

Parameter Details

Regex Pattern Examples

Response

Success Response

Response Fields

Code Examples

cURL Examples

JavaScript/Node.js

TypeScript

Python

Common Use Cases

1. Search Bar Implementation

3. Category-Based Search

4. Advanced Search with Multiple Criteria

Regex Tips & Tricks

Common Patterns

URL Encoding

Best Practices

Search Patterns

Performance

User Experience

Error Handling

Limitations

Back: List Specs

Authorizations

Query Parameters

Response

Using the API

Spec Generation

Spec Management

​GET /find-specs

​Overview

​Endpoint

​Headers

​Query Parameters

​Parameter Details

​Regex Pattern Examples

​Response

​Success Response

​Response Fields

​Code Examples

​cURL Examples

​JavaScript/Node.js

​TypeScript

​Python

​Common Use Cases

​1. Search Bar Implementation

​2. Find Related Specs

​3. Category-Based Search

​4. Advanced Search with Multiple Criteria

​Regex Tips & Tricks

​Common Patterns

​URL Encoding

​Best Practices

​Search Patterns

​Performance

​User Experience

​Error Handling

​Limitations

Back: List Specs

Authorizations

Query Parameters

Response

GET /find-specs

Overview

Endpoint

Headers

Query Parameters

Parameter Details

Regex Pattern Examples

Response

Success Response

Response Fields

Code Examples

cURL Examples

JavaScript/Node.js

TypeScript

Python

Common Use Cases

1. Search Bar Implementation

2. Find Related Specs

3. Category-Based Search

4. Advanced Search with Multiple Criteria

Regex Tips & Tricks

Common Patterns

URL Encoding

Best Practices

Search Patterns

Performance

User Experience

Error Handling

Limitations