Skip to content

slugkit/slugkit-py-sdk

BranchesTags

Repository files navigation

SlugKit Python SDK

A Python SDK for generating human-readable IDs using SlugKit.dev service.

Please see SlugKit documentation for information on creating series and obtaining API keys.

Installation

pip install slugkit-py-sdk

API Overview

The SDK provides a comprehensive interface to generate human-readable IDs, test patterns, and manage generator state.

Basic Usage

from slugkit import SyncClient

# Initialize the client
client = SyncClient(
    base_url="https://dev.slugkit.dev/api/v1",
    api_key="your-api-key"
)

# Generate a single ID
id = client.series.mint()[0]

# Generate multiple IDs
ids = client.series.mint(count=5)

# Get generator stats
stats = client.series.stats()

# Get series information
series_info = client.series.info()

# Get list of available series
series_list = client.series.list()

# Get subscription limits and features
limits = client.limits()

# Reset the generator
client.series.reset()

Advanced Generator Configuration

# Configure generator with limits and batch sizes
generator = client.series.mint.with_limit(100).with_batch_size(10)

# Start from a specific sequence
generator = generator.starting_from(1000)

# Use dry-run mode (for testing)
generator = generator.with_dry_run()

# Generate IDs with configuration
ids = list(generator)

Pattern Testing (Forge)

# Test a pattern with various options
ids = client.forge(
    pattern="{adjective}-{noun}-{number:3d}",
    seed="optional-seed",
    sequence=1,
    count=5
)

# Get pattern information
pattern_info = client.forge.pattern_info("{adjective}-{noun}-{number:3d}")

# Get dictionary information
dict_info = client.forge.dictionary_info()

# Get dictionary tags
dict_tags = client.forge.dictionary_tags()

# Use different pattern types
ids = client.forge(
    pattern="simple-{noun}-{number:2,hex}",
    count=3
)

Series Management

# Access a specific series
series_generator = client.series["series-slug"]

# Generate IDs for specific series
ids = series_generator(count=10)

# Get stats for specific series
stats = series_generator.stats()

# Get series info for specific series
series_info = series_generator.series_info()

Async Usage

The SDK also provides an async client for use with async/await code:

import asyncio
from slugkit import AsyncClient

async def main():
    # Initialize the async client
    client = AsyncClient(
        base_url="https://dev.slugkit.dev/api/v1",
        api_key="your-api-key"
    )

    # Generate a single ID
    id = await client.series.mint()[0]

    # Generate multiple IDs
    ids = await client.series.mint(count=5)

    # Get generator stats
    stats = await client.series.stats()

    # Get series information
    series_info = await client.series.info()

    # Get list of available series
    series_list = await client.series.list()

    # Get subscription limits and features
    limits = await client.limits()

    # Reset the generator
    await client.series.reset()

    # Test a pattern
ids = await client.forge(
    pattern="{adjective}-{noun}-{number:3d}",
    seed="optional-seed",
    sequence=1,
    count=5
)

# Get pattern information
pattern_info = await client.forge.pattern_info("{adjective}-{noun}-{number:3d}")

# Get dictionary information
dict_info = await client.forge.dictionary_info()

# Get dictionary tags
dict_tags = await client.forge.dictionary_tags()

    # Stream IDs asynchronously
    async for id in client.series.mint:
        print(id)

# Run the async code
asyncio.run(main())

Command Line Interface

The SDK includes a command-line interface for easy usage:

Generate IDs

# Generate a single ID
slugkit mint

# Generate multiple IDs
slugkit mint 5

# Generate IDs with custom batch size
slugkit mint 10 --batch-size 2

Test Patterns

# Test a pattern
slugkit forge "your-pattern"

# Test with specific seed and sequence
slugkit forge "your-pattern" --seed "my-seed" --sequence 1

# Generate multiple test IDs
slugkit forge "your-pattern" --count 5

# Validate a pattern and get information
slugkit validate "your-pattern"

Generator Management

# Get generator stats
slugkit stats

# Get list of available series
slugkit series-list

# Get series information
slugkit series-info

# Get subscription limits and features
slugkit limits

# Get subscription limits in JSON format
slugkit limits --output-format json

# Reset the generator
slugkit reset

Configuration

The CLI can be configured using environment variables or command-line options:

# Set base URL
export SLUGKIT_BASE_URL="https://dev.slugkit.dev/api/v1"

# Set API key
export SLUGKIT_API_KEY="your-api-key"

# Or use command-line options
slugkit --base-url "https://dev.slugkit.dev/api/v1" --api-key "your-api-key" mint

Output Format

The CLI supports different output formats:

# Text output (default)
slugkit mint

# JSON output
slugkit mint --output-format json

Data Types

The SDK provides structured data types for API responses:

StatsItem

Represents generator statistics:

from slugkit.base import StatsItem

stats = client.series.stats()
for item in stats:
    print(f"Event: {item.event_type}")
    print(f"Total Count: {item.total_count}")
    print(f"Request Count: {item.request_count}")
    print(f"Average Duration: {item.avg_duration_us}μs")

SeriesInfo

Represents series information:

from slugkit.base import SeriesInfo

series_info = client.series.info()
print(f"Pattern: {series_info.pattern}")
print(f"Capacity: {series_info.capacity}")
print(f"Generated: {series_info.generated_count}")
print(f"Last Modified: {series_info.mtime}")

PatternInfo

Represents pattern information:

from slugkit.base import PatternInfo

pattern_info = client.forge.pattern_info("test-{adjective}-{noun}")
print(f"Pattern: {pattern_info.pattern}")
print(f"Capacity: {pattern_info.capacity}")
print(f"Max Slug Length: {pattern_info.max_slug_length}")
print(f"Complexity: {pattern_info.complexity}")
print(f"Components: {pattern_info.components}")

DictionaryInfo

Represents dictionary information:

from slugkit.base import DictionaryInfo

dict_info = client.forge.dictionary_info()
for item in dict_info:
    print(f"Kind: {item.kind}, Count: {item.count}")

DictionaryTag

Represents dictionary tags:

from slugkit.base import DictionaryTag

dict_tags = client.forge.dictionary_tags()
for tag in dict_tags:
    print(f"Kind: {tag.kind}, Tag: {tag.tag}, Words: {tag.word_count}")

SubscriptionFeatures

Represents subscription limits and features:

from slugkit.base import SubscriptionFeatures

limits = client.limits()
print(f"Max Series: {limits.max_series}")
print(f"Requests Per Minute: {limits.req_per_minute}")
print(f"Forge Enabled: {limits.forge_enabled}")
print(f"Max Mint Per Day: {limits.max_mint_per_day}")
print(f"Max Mint Per Month: {limits.max_mint_per_month}")

# All fields are optional - None means not available
if limits.max_forge_per_request is not None:
    print(f"Max Forge Per Request: {limits.max_forge_per_request}")

API Response Examples

Stats Response

[
  {
    "event_type": "mint",
    "date_part": "total",
    "total_count": 104517,
    "request_count": 118,
    "total_duration_us": 1092348,
    "avg_duration_us": 10.45
  },
  {
    "event_type": "forge",
    "date_part": "total",
    "total_count": 15,
    "request_count": 5,
    "total_duration_us": 500,
    "avg_duration_us": 50.0
  }
]

Series Info Response

{
  "slug": "whole-blond-rower-a597",
  "org_slug": "alias-first-glute-67d9",
  "pattern": "{adverb}-{adjective}-{noun}-{number:3d}",
  "max_pattern_length": 80,
  "capacity": "1281739952493000",
  "generated_count": "105",
  "mtime": "2025-08-29T00:54:35.128902+00:00"
}

Pattern Info Response

{
  "pattern": "{adjective}-{noun}-{number:3d}",
  "capacity": "1000000",
  "max_slug_length": 25,
  "complexity": 3,
  "components": 3
}

Dictionary Info Response

[
  {
    "kind": "adjective",
    "count": 1000
  },
  {
    "kind": "noun",
    "count": 2000
  },
  {
    "kind": "verb",
    "count": 1500
  }
]

Dictionary Tags Response

[
  {
    "kind": "adjective",
    "tag": "positive",
    "description": "Positive adjectives",
    "opt_in": true,
    "word_count": 500
  },
  {
    "kind": "noun",
    "tag": "animals",
    "description": "Animal names",
    "opt_in": false,
    "word_count": 300
  }
]

Series List Response

[
  "series-1",
  "series-2",
  "series-3"
]

Error Handling

The SDK provides comprehensive error handling with custom exception types:

from slugkit import SyncClient
from slugkit.base import (
    SlugKitConnectionError,
    SlugKitAuthenticationError,
    SlugKitValidationError,
    SlugKitRateLimitError,
    SlugKitQuotaError,
    SlugKitServerError,
)

client = SyncClient(
    base_url="https://dev.slugkit.dev/api/v1",
    api_key="your-api-key"
)

try:
    ids = client.series.mint(count=10)
except SlugKitRateLimitError as e:
    # Rate limiting with detailed information
    print(f"Rate limited: {e}")
    if e.rate_limit_reason:
        print(f"Reason: {e.rate_limit_reason}")
    if e.retry_after:
        print(f"Retry after: {e.retry_after} seconds")
    if e.rpm_remaining is not None:
        print(f"RPM remaining: {e.rpm_remaining}")
    if e.daily_remaining is not None:
        print(f"Daily remaining: {e.daily_remaining}")
except SlugKitAuthenticationError as e:
    print(f"Authentication failed: {e}")
except SlugKitValidationError as e:
    print(f"Validation error: {e}")
except SlugKitQuotaError as e:
    print(f"Quota exceeded: {e}")
except SlugKitConnectionError as e:
    print(f"Connection failed: {e}")
except SlugKitServerError as e:
    print(f"Server error: {e}")

Rate Limiting

The SDK implements intelligent rate limit handling:

  • Automatic Retry: Rate limit errors with rate-limit-exceeded or daily-limit-exceeded are automatically retried
  • Exponential Backoff: Retry delays increase exponentially with jitter
  • Server Retry-After: Respects X-Slug-Retry-After header from the server
  • Quota Visibility: Shows remaining RPM, daily, monthly, and lifetime quotas
  • Smart Non-Retry: Permanent failures (not-available, request-size-exceeded) and long-term limits (monthly-limit-exceeded, lifetime-limit-exceeded) are not retried

Rate limit reasons returned by the server:

  • rate-limit-exceeded - RPM limit reached (retryable with accurate retry-after)
  • daily-limit-exceeded - Daily limit reached (retryable with accurate retry-after)
  • monthly-limit-exceeded - Monthly quota exceeded (non-retryable, wait until next month)
  • lifetime-limit-exceeded - Lifetime quota exceeded (non-retryable, upgrade required)
  • request-size-exceeded - Request size too large (non-retryable, reduce batch size)
  • not-available - Feature not available to user (non-retryable, check subscription)
  • redis-error - Server error (non-retryable, contact support)

Pattern Language

The SDK supports SlugKit's pattern language for generating structured IDs:

Basic Tokens

  • {adjective} - Random adjective
  • {noun} - Random noun
  • {adverb} - Random adverb
  • {verb} - Random verb

Number Generators

  • {number:3d} - 3-digit decimal number
  • {number:4,hex} - 4-character hexadecimal
  • {number:2r} - up to 2-character roman number, lowercase
  • {number:6R} - up to 6-character roman number, uppercase

Examples

# Simple patterns
pattern1 = "{adjective}-{noun}"
pattern2 = "{color}-{animal}-{number:3d}"

# Complex patterns with constraints
pattern3 = "{adverb}-{adjective}-{noun}-{number:4,hex}[==5]"

# Generate IDs
ids = client.forge(pattern=pattern3, count=5)

MCP Server: AI Assistant Integration

⚡ The SlugKit MCP Server enables AI assistants to generate truly unique identifiers!

The Problem with AI-Generated IDs

AI assistants cannot generate truly unique strings because they:

  • Have no persistence between conversations
  • Cannot coordinate across multiple instances
  • Generate probabilistic, not truly random content
  • Have no concept of sequential uniqueness

This makes AI-generated identifiers unsuitable for production use where uniqueness is critical.

The Solution: SlugKit MCP Integration

The SlugKit MCP (Model Context Protocol) server connects AI assistants like Claude Desktop, Cursor, and Warp directly to SlugKit's guaranteed-unique ID generation service.

Key Benefits:

  • ✅ Guaranteed Uniqueness - Mathematical guarantees, not probabilistic generation
  • ✅ Sequential Consistency - Proper sequential IDs across all sessions
  • ✅ Production Ready - Battle-tested for high-volume applications
  • ✅ Zero Collisions - Impossible to generate duplicate IDs
  • ✅ Pattern Flexibility - Generate IDs matching your exact requirements

Quick Setup

Cloud Instance (Recommended):

{
    "mcpServers": {
        "SlugKit": {
            "command": "npx",
            "args": [
                "-y", "mcp-remote",
                "https://dev.slugkit.dev/api/v1/mcp",
                "--header", "x-api-key:YOUR_API_KEY_HERE"
            ]
        }
    }
}

Local Installation:

pip install slugkit-py-sdk
{
    "mcpServers": {
        "SlugKit": {
            "command": "slugkit-mcp",
            "env": {
                "SLUGKIT_API_KEY": "YOUR_API_KEY_HERE"
            }
        }
    }
}

Usage Examples

Once configured, ask your AI assistant:

"Generate 10 unique user IDs with the pattern user-{adjective}-{noun}-{number:4d}"

→ Returns guaranteed-unique IDs like "user-happy-dolphin-0001", "user-bright-falcon-0002"

"Analyze this pattern: {color}-{animal}-{number:3,hex} - how many unique IDs can it generate?"

→ Shows capacity analysis: "1.2M unique combinations, medium complexity"

"Compare these URL slug patterns and recommend the best one: {adjective}-{noun} vs {verb}-{noun}-{number:2d}"

→ Generates samples, analyzes capacity, recommends optimal choice

Available Capabilities

The MCP server provides AI assistants with access to:

  • ID Generation: forge (custom patterns), mint (series), slice (preview)
  • Pattern Analysis: validate_pattern, analyze_pattern, compare_patterns
  • Information: dictionary_info, series_info, key_info, stats
  • Management: series_list, reset (dev/test)
  • Documentation: get_help_topic, pattern syntax references

📚 For detailed setup instructions, troubleshooting, and advanced configuration, see README_MCP.md

License

This project is licensed under the MIT License - see the LICENSE file for details.

About

Python SDK for SlugKit.dev

Resources

Readme

License

MIT license
Activity

Stars

0 stars

Watchers

0 watching

Forks

0 forks
Report repository

Releases

No releases published

Packages

No packages published

Contributors 3

  •  
  •  
  •  

Languages