Skip to main content

Official Python client for PLP (Prompt Library Protocol)

Project description

plp-client

PLP Compliant PyPI version License: MIT

Official Python client for PLP (Prompt Library Protocol).

Installation

pip install plp-client

Quick Start

from plp_client import PLPClient, PromptInput

# Initialize client
client = PLPClient("https://prompts.goreal.ai", api_key="your-api-key")

# Get latest version of a prompt
prompt = client.get("marketing/welcome-email")
print(prompt.content)  # "Hello {{name}}..."

# Get specific version
old_prompt = client.get("marketing/welcome-email", "1.0.0")

# Save a new prompt
client.put("support/faq-bot", PromptInput(
    content="You are a helpful FAQ bot. Answer: {{question}}",
    meta={"version": "1.0.0", "author": "yoad"}
))

# Delete a prompt
client.delete("deprecated/old-prompt")

API Reference

PLPClient

Constructor

PLPClient(base_url: str, api_key: Optional[str] = None, 
          headers: Optional[Dict[str, str]] = None, timeout: int = 10)

Parameters:

  • base_url (str): Base URL of the PLP server
  • api_key (str, optional): Optional Bearer token for authentication
  • headers (dict, optional): Additional HTTP headers
  • timeout (int): Request timeout in seconds (default: 10)

Methods

get(prompt_id, version=None)

Retrieve a prompt by ID and optional version.

def get(prompt_id: str, version: Optional[str] = None) -> PromptEnvelope
  • prompt_id (str): Unique prompt identifier (e.g., "marketing/welcome-email")
  • version (str, optional): Optional version string (e.g., "1.2.0"). If omitted, returns latest.

Returns: PromptEnvelope

Raises: PLPError if not found (404) or other errors

put(prompt_id, input)

Create or update a prompt (idempotent upsert).

def put(prompt_id: str, input: PromptInput) -> PromptEnvelope
  • prompt_id (str): Unique prompt identifier
  • input (PromptInput): Object with content and optional meta

Returns: The saved PromptEnvelope

delete(prompt_id)

Delete a prompt and all its versions.

def delete(prompt_id: str) -> None
  • prompt_id (str): Unique prompt identifier

Returns: None (204 No Content on success)

Aliases
fetch(prompt_id, version=None)  # Alias for get()
save(prompt_id, input)          # Alias for put()

Data Classes

PromptEnvelope

class PromptEnvelope:
    id: str
    content: str
    meta: Dict[str, Any]

PromptInput

class PromptInput:
    content: str
    meta: Dict[str, Any]

Error Handling

from plp_client import PLPClient, PLPError

client = PLPClient("https://prompts.goreal.ai")

try:
    prompt = client.get("missing/prompt")
except PLPError as e:
    print(f"Error: {e}")
    print(f"Status code: {e.status_code}")

Examples

Using with OpenAI

from plp_client import PLPClient
from openai import OpenAI

plp = PLPClient("https://prompts.goreal.ai")
openai = OpenAI()

prompt = plp.get("assistant/code-reviewer")

response = openai.chat.completions.create(
    model="gpt-4",
    messages=[{"role": "user", "content": prompt.content}],
    temperature=prompt.meta.get("model_config", {}).get("temperature", 0.7)
)

print(response.choices[0].message.content)

Using with Anthropic

from plp_client import PLPClient
import anthropic

plp = PLPClient("https://prompts.goreal.ai")
client = anthropic.Anthropic()

prompt = plp.get("creative/story-generator")

message = client.messages.create(
    model="claude-3-opus-20240229",
    max_tokens=prompt.meta.get("model_config", {}).get("max_tokens", 1024),
    messages=[{"role": "user", "content": prompt.content}]
)

print(message.content)

Context Manager

from plp_client import PLPClient

with PLPClient("https://prompts.goreal.ai") as client:
    prompt = client.get("my/prompt")
    print(prompt.content)
# Session is automatically closed

Development

# Install dependencies
pip install -e ".[dev]"

# Run tests
pytest

# Run tests with coverage
pytest --cov=plp_client

# Format code
black src/ tests/

# Type checking
mypy src/

License

MIT © GoReal.AI

Contributing

See CONTRIBUTING.md

Learn More

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

plp_client-1.0.0.tar.gz (8.0 kB view details)

Uploaded Source

Built Distribution

If you're not sure about the file name format, learn more about wheel file names.

plp_client-1.0.0-py3-none-any.whl (6.3 kB view details)

Uploaded Python 3

File details

Details for the file plp_client-1.0.0.tar.gz.

File metadata

  • Download URL: plp_client-1.0.0.tar.gz
  • Upload date:
  • Size: 8.0 kB
  • Tags: Source
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/6.2.0 CPython/3.11.14

File hashes

Hashes for plp_client-1.0.0.tar.gz
Algorithm Hash digest
SHA256 99ae8b28bdb0085828f11a09a8a6786a087ad818b5e119511c0845d40571e877
MD5 55f9b9948a457aab0ba7d1b8d44c6a4d
BLAKE2b-256 f037a8ce39198a52f6965a8c4c829d43e7b2a3e7097bc77c56a57f97b086b541

See more details on using hashes here.

File details

Details for the file plp_client-1.0.0-py3-none-any.whl.

File metadata

  • Download URL: plp_client-1.0.0-py3-none-any.whl
  • Upload date:
  • Size: 6.3 kB
  • Tags: Python 3
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/6.2.0 CPython/3.11.14

File hashes

Hashes for plp_client-1.0.0-py3-none-any.whl
Algorithm Hash digest
SHA256 601c6b897f470b278b80c4a04eaee2663925ecb65d2af9e2f56af75a69682447
MD5 0d800024408c73e7190088d3f3325a4a
BLAKE2b-256 b0aee3e723c99537c55cbf9e2730e05985ef9bd2a2fc5e7d929821c0696f1045

See more details on using hashes here.

Supported by

AWS Cloud computing and Security Sponsor Datadog Monitoring Depot Continuous Integration Fastly CDN Google Download Analytics Pingdom Monitoring Sentry Error logging StatusPage Status page