Name | promptl-ai JSON |
Version |
0.1.0b1
JSON |
| download |
home_page | None |
Summary | PromptL bindings for Python |
upload_time | 2025-01-22 16:33:57 |
maintainer | None |
docs_url | None |
author | None |
requires_python | <3.13,>=3.9 |
license | None |
keywords |
|
VCS |
|
bugtrack_url |
|
requirements |
No requirements were recorded.
|
Travis-CI |
No Travis.
|
coveralls test coverage |
No coveralls.
|
# PromptL
```sh
pip install promptl-ai
```
Requires Python `3.9` or higher.
## What is PromptL?
[PromptL](https://promptl.ai/) offers a common, easy-to-use syntax for defining dynamic prompts for LLMs. It is a simple, yet powerful language that allows you to define prompts in a human-readable format, while still being able to leverage the full power of LLMs.
> This repository contains the Python bindings for PromptL. Go to the main [repository](https://github.com/latitude-dev/promptl) to learn more.
## Usage
```python
from promptl_ai import Promptl
promptl = Promptl()
prompt = """
<step>
First, think step by step about how to answer the user's question.
<user>
Taking into account this context: {{context}}
I have the following question: {{question}}
</user>
</step>
<step>
Finally, answer the user's question succinctly yet complete.
</step>
"""
chain = promptl.chains.create(
prompt=prompt,
parameters={
"context": "PromptL is a templating language specifically designed for LLM prompting.",
"question": "What is PromptL?",
},
)
conversation = chain.step()
conversation = chain.step("Reasoning...") # LLM response
conversation = chain.step("Answer...") # LLM response
print(conversation.messages)
```
Find more [examples](examples).
## Development
Requires uv `0.5.10` or higher.
- Install dependencies: `uv venv && uv sync --all-extras --all-groups`
- Add [dev] dependencies: `uv add [--dev] <package>`
- Run linter: `uv run scripts/lint.py`
- Run formatter: `uv run scripts/format.py`
- Run tests: `uv run scripts/test.py`
- Build package: `uv build`
- Publish package: `uv publish`
## License
The PromptL bindings are licensed under the [MIT License](https://opensource.org/licenses/MIT) - read the [LICENSE](LICENSE) file for details.
Raw data
{
"_id": null,
"home_page": null,
"name": "promptl-ai",
"maintainer": null,
"docs_url": null,
"requires_python": "<3.13,>=3.9",
"maintainer_email": "Latitude Data SL <hello@latitude.so>",
"keywords": null,
"author": null,
"author_email": "Latitude Data SL <hello@latitude.so>",
"download_url": "https://files.pythonhosted.org/packages/75/92/9731f131373b5fa63a7f9908aab0b19e65dcd00a1dda4777e415f274b085/promptl_ai-0.1.0b1.tar.gz",
"platform": null,
"description": "# PromptL\n\n```sh\npip install promptl-ai\n```\n\nRequires Python `3.9` or higher.\n\n## What is PromptL?\n\n[PromptL](https://promptl.ai/) offers a common, easy-to-use syntax for defining dynamic prompts for LLMs. It is a simple, yet powerful language that allows you to define prompts in a human-readable format, while still being able to leverage the full power of LLMs.\n\n> This repository contains the Python bindings for PromptL. Go to the main [repository](https://github.com/latitude-dev/promptl) to learn more.\n\n## Usage\n\n```python\nfrom promptl_ai import Promptl\n\npromptl = Promptl()\n\nprompt = \"\"\"\n<step>\n First, think step by step about how to answer the user's question.\n <user>\n Taking into account this context: {{context}}\n I have the following question: {{question}}\n </user>\n</step>\n<step>\n Finally, answer the user's question succinctly yet complete.\n</step>\n\"\"\"\n\nchain = promptl.chains.create(\n prompt=prompt,\n parameters={\n \"context\": \"PromptL is a templating language specifically designed for LLM prompting.\",\n \"question\": \"What is PromptL?\",\n },\n)\nconversation = chain.step()\nconversation = chain.step(\"Reasoning...\") # LLM response\nconversation = chain.step(\"Answer...\") # LLM response\n\nprint(conversation.messages)\n```\n\nFind more [examples](examples).\n\n## Development\n\nRequires uv `0.5.10` or higher.\n\n- Install dependencies: `uv venv && uv sync --all-extras --all-groups`\n- Add [dev] dependencies: `uv add [--dev] <package>`\n- Run linter: `uv run scripts/lint.py`\n- Run formatter: `uv run scripts/format.py`\n- Run tests: `uv run scripts/test.py`\n- Build package: `uv build`\n- Publish package: `uv publish`\n\n## License\n\nThe PromptL bindings are licensed under the [MIT License](https://opensource.org/licenses/MIT) - read the [LICENSE](LICENSE) file for details.\n",
"bugtrack_url": null,
"license": null,
"summary": "PromptL bindings for Python",
"version": "0.1.0b1",
"project_urls": {
"documentation": "https://docs.latitude.so/promptl",
"homepage": "https://github.com/latitude-dev/promptl-py#readme",
"repository": "https://github.com/latitude-dev/promptl-py"
},
"split_keywords": [],
"urls": [
{
"comment_text": null,
"digests": {
"blake2b_256": "51edb3baa40257769b4f6cbffe4f65a88cfa79a4507908bed409ccffed560a7b",
"md5": "36a1a16cc00478e9d56746a43c9a48da",
"sha256": "ca30156635b85a18470bb0369ed436f7c6f248e780a883f3469c7388bb097f25"
},
"downloads": -1,
"filename": "promptl_ai-0.1.0b1-py3-none-any.whl",
"has_sig": false,
"md5_digest": "36a1a16cc00478e9d56746a43c9a48da",
"packagetype": "bdist_wheel",
"python_version": "py3",
"requires_python": "<3.13,>=3.9",
"size": 964276,
"upload_time": "2025-01-22T16:33:54",
"upload_time_iso_8601": "2025-01-22T16:33:54.825050Z",
"url": "https://files.pythonhosted.org/packages/51/ed/b3baa40257769b4f6cbffe4f65a88cfa79a4507908bed409ccffed560a7b/promptl_ai-0.1.0b1-py3-none-any.whl",
"yanked": false,
"yanked_reason": null
},
{
"comment_text": null,
"digests": {
"blake2b_256": "75929731f131373b5fa63a7f9908aab0b19e65dcd00a1dda4777e415f274b085",
"md5": "34da25dc3ab7f105e3951d1198e0feb8",
"sha256": "5594961f0fd390a2028ae618b385a3441518fffc6d78943d14d892bf19ec8f1d"
},
"downloads": -1,
"filename": "promptl_ai-0.1.0b1.tar.gz",
"has_sig": false,
"md5_digest": "34da25dc3ab7f105e3951d1198e0feb8",
"packagetype": "sdist",
"python_version": "source",
"requires_python": "<3.13,>=3.9",
"size": 977158,
"upload_time": "2025-01-22T16:33:57",
"upload_time_iso_8601": "2025-01-22T16:33:57.874908Z",
"url": "https://files.pythonhosted.org/packages/75/92/9731f131373b5fa63a7f9908aab0b19e65dcd00a1dda4777e415f274b085/promptl_ai-0.1.0b1.tar.gz",
"yanked": false,
"yanked_reason": null
}
],
"upload_time": "2025-01-22 16:33:57",
"github": true,
"gitlab": false,
"bitbucket": false,
"codeberg": false,
"github_user": "latitude-dev",
"github_project": "promptl-py#readme",
"travis_ci": false,
"coveralls": false,
"github_actions": true,
"lcname": "promptl-ai"
}