01 Skills Create
Function: create¶
Generates a CodeSkill instance using different input sources.
Parameters:¶
request: String detailing the skill functionality.messagesormessages_json_path: Messages as a list of dictionaries or a path to a JSON file containing messages.file_contentorfile_path: String of file content or path to a code/API doc file.skill_pathorskill_json_path: Directory path with skill name as stem or file path withskill.jsonas stem.
Returns:¶
CodeSkill: The created skill.
Usage:¶
- Creating Skill using a Request String:
skill = create(request="filter how many prime numbers are in 201")
In [1]:
Copied!
from creator import create
from creator import create
The fundamental principle of the "create API" is to utilize various input sources to assemble messages. These messages are then forwarded to a tagging agent to construct a consistent skill object.
In [2]:
Copied!
?create
?create
Signature: create( request: Optional[str] = None, messages: Optional[List[dict]] = None, messages_json_path: Optional[str] = None, skill_path: Optional[str] = None, skill_json_path: Optional[str] = None, file_content: Optional[str] = None, file_path: Optional[str] = None, huggingface_repo_id: Optional[str] = None, huggingface_skill_path: Optional[str] = None, save: bool = False, ) -> creator.core.skill.CodeSkill Docstring: Main method to create a new skill. File: ~/miniconda3/envs/open_creator_online/lib/python3.10/site-packages/creator/core/core.py Type: method
1. Creating a Skill Using a Request String¶
The request will be forwarded to a code interpreter agent, which will facilitate multi-turn problem-solving experiences. This process will generate a sequence of messages.
These messages will then be relayed to a skill extractor agent, which will produce a formatted skill object.
In [3]:
Copied!
skill = create(request="filter how many prime numbers are in 201")
skill = create(request="filter how many prime numbers are in 201")
Output()
Output()
Output()
Output()
Output()
Output()
In [4]:
Copied!
skill.show()
skill.show()
Skill Details:
• Name: count_prime_numbers
• Description: This skill counts the number of prime numbers within a given range. A prime number is a natural
number greater than 1 that has no positive divisors other than 1 and itself.
• Version: 1.0.0
• Usage:
count_prime_numbers(201) # returns 46
• Parameters:
• n (integer): The upper limit of the range within which to count prime numbers.
• Required: True
• Returns:
• prime_count (integer): The number of prime numbers within the given range.
2 Creating Skill using Messages¶
This only use case is for the skill extractor agent. The messages are forwarded to the skill extractor agent, which will produce a formatted skill object.
In [5]:
Copied!
messages = [
{
"role": "user",
"content": "write a program to list all the python functions and their docstrings in a directory"
},
{
"role": "assistant",
"content": "Sure, I can help with that. Here's the plan:\n\n1. First, we need to get a list of all Python files in the specified directory. We can do this by using the `os` and `glob` modules in Python.\n2. Then, for each Python file, we will parse the file to find all function definitions. We can do this by using the `ast` module in Python, which can parse Python source code into an abstract syntax tree (AST).\n3. For each function definition, we will extract the function's name and its docstring. The `ast` module can also help us with this.\n4. Finally, we will print out the function names and their docstrings.\n\nLet's start with step 1: getting a list of all Python files in the specified directory.",
"function_call": {
"name": "run_code",
"arguments": "{\n \"language\": \"python\",\n \"code\": \"import os\\nimport glob\\n\\n# Get the current working directory\\ncwd = os.getcwd()\\n\\n# Get a list of all Python files in the directory\\npython_files = glob.glob(os.path.join(cwd, '*.py'))\\n\\npython_files\"\n}",
}
}
]
messages = [
{
"role": "user",
"content": "write a program to list all the python functions and their docstrings in a directory"
},
{
"role": "assistant",
"content": "Sure, I can help with that. Here's the plan:\n\n1. First, we need to get a list of all Python files in the specified directory. We can do this by using the `os` and `glob` modules in Python.\n2. Then, for each Python file, we will parse the file to find all function definitions. We can do this by using the `ast` module in Python, which can parse Python source code into an abstract syntax tree (AST).\n3. For each function definition, we will extract the function's name and its docstring. The `ast` module can also help us with this.\n4. Finally, we will print out the function names and their docstrings.\n\nLet's start with step 1: getting a list of all Python files in the specified directory.",
"function_call": {
"name": "run_code",
"arguments": "{\n \"language\": \"python\",\n \"code\": \"import os\\nimport glob\\n\\n# Get the current working directory\\ncwd = os.getcwd()\\n\\n# Get a list of all Python files in the directory\\npython_files = glob.glob(os.path.join(cwd, '*.py'))\\n\\npython_files\"\n}",
}
}
]
In [6]:
Copied!
skill = create(messages=messages)
skill = create(messages=messages)
In [7]:
Copied!
skill.show()
skill.show()
Skill Details:
• Name: list_python_functions_and_docstrings
• Description: This skill lists all the Python functions and their docstrings in a specified directory. It first
gets a list of all Python files in the directory, then parses each file to find all function definitions. For
each function definition, it extracts the function's name and its docstring.
• Version: 1.0.0
• Usage:
list_python_functions_and_docstrings('/path/to/directory')
• Parameters:
• directory (string): The directory to search for Python files.
• Required: True
• Returns:
Also, you can pass a messages JSON file path to the messages_json_path parameter.
In [8]:
Copied!
skill = create(messages_json_path="./data/messages_example.json")
skill = create(messages_json_path="./data/messages_example.json")
In [9]:
Copied!
skill.show()
skill.show()
Skill Details:
• Name: extract_pages_from_pdf
• Description: This skill extracts a specified range of pages from a PDF file and saves them as a new PDF file.
The user needs to provide the path to the original PDF file and the range of pages to be extracted. The
extracted pages are saved in a new PDF file in the current working directory.
• Version: 1.0.0
• Usage:
extract_pages_from_pdf('~/Downloads/voyager.pdf', 2, 5, 'voyager_extracted.pdf')
• Parameters:
• pdf_path (string): Path to the original PDF file.
• Required: True
• start_page (integer): The first page to be extracted. Page numbers start from 1.
• Required: True
• end_page (integer): The last page to be extracted. This page is included in the extraction.
• Required: True
• output_file (string): Name of the output file where the extracted pages will be saved.
• Default: 'extracted_pages.pdf'
• Returns:
• output_file (string): Name of the output file where the extracted pages were saved.
3. Creating Skill using a File Content (code, API doc, etc.)¶
Different to create a skill by request, we construct the messages by parsing the file content. The messages are then forwarded to the skill extractor agent, which will produce a formatted skill object.
In [10]:
Copied!
code_file_content = """
from rich import print as rich_print
from rich.markdown import Markdown
from rich.rule import Rule
def display_markdown_message(message):
'''Display markdown message. Works with multiline strings with lots of indentation.
Will automatically make single line > tags beautiful.
'''
for line in message.split("\n"):
line = line.strip()
if line == "":
print("")
elif line == "---":
rich_print(Rule(style="white"))
else:
rich_print(Markdown(line))
if "\n" not in message and message.startswith(">"):
# Aesthetic choice. For these tags, they need a space below them
print("")
"""
code_file_content = """
from rich import print as rich_print
from rich.markdown import Markdown
from rich.rule import Rule
def display_markdown_message(message):
'''Display markdown message. Works with multiline strings with lots of indentation.
Will automatically make single line > tags beautiful.
'''
for line in message.split("\n"):
line = line.strip()
if line == "":
print("")
elif line == "---":
rich_print(Rule(style="white"))
else:
rich_print(Markdown(line))
if "\n" not in message and message.startswith(">"):
# Aesthetic choice. For these tags, they need a space below them
print("")
"""
In [11]:
Copied!
skill = create(file_content=code_file_content)
skill = create(file_content=code_file_content)
In [12]:
Copied!
skill.show()
skill.show()
Skill Details:
• Name: display_markdown_message
• Description: This skill is used to display a markdown message. It works with multiline strings with lots of
indentation and will automatically make single line > tags beautiful.
• Version: 1.0.0
• Usage:
display_markdown_message('> This is a markdown message.')
• Parameters:
• message (string): The markdown message to be displayed.
• Required: True
• Returns:
Also, you can pass a API doc to the file_path parameter.
In [13]:
Copied!
skill = create(file_path="./data/create_api.md")
skill = create(file_path="./data/create_api.md")
In [14]:
Copied!
skill.show()
skill.show()
Skill Details:
• Name: create
• Description: This function generates a CodeSkill instance using different input sources. It can take in a
request string detailing the skill functionality, messages or a path to a JSON file containing messages, a
string of file content or path to a code/API doc file, a directory path with skill name as stem or file path
with skill.json as stem, an identifier for a Huggingface repository, or a path to the skill within the
Huggingface repository.
• Version: 1.0.0
• Usage:
from creator import create
skill = create(request='...', messages=[...], file_content='...', file_path='...', skill_path='...',
skill_json_path='...', huggingface_repo_id='...', huggingface_skill_path='...')
• Parameters:
• request (string): String detailing the skill functionality.
• messages (array): Messages as a list of dictionaries.
• messages_json_path (string): Path to a JSON file containing messages.
• file_content (string): String of file content.
• file_path (string): Path to a code/API doc file.
• skill_path (string): Directory path with skill name as stem.
• skill_json_path (string): File path with skill.json as stem.
• huggingface_repo_id (string): Identifier for a Huggingface repository.
• huggingface_skill_path (string): Path to the skill within the Huggingface repository.
• Returns:
• CodeSkill (object): The created skill.
4. Load a created skill from skill library¶
You can load a skill from the skill library by passing the skill path to the skill_path parameter. The skill path can be a directory path with skill name as stem or file path with skill.json as stem.
The default skill library path is ~/.cache/open-creator/skill_library.
In [15]:
Copied!
# or
# from creator import save
# save(skill)
skill.save()
# or
# from creator import save
# save(skill)
skill.save()
▌ saved to /Users/gongjunmin/.cache/open_creator/skill_library/create
In [16]:
Copied!
skill = create(skill_path="/Users/gongjunmin/.cache/open_creator/skill_library/create_api")
# or
# skill = create(skill_path="/Users/gongjunmin/.cache/open_creator/skill_library/create_api/skill.json")
skill.show()
skill = create(skill_path="/Users/gongjunmin/.cache/open_creator/skill_library/create_api")
# or
# skill = create(skill_path="/Users/gongjunmin/.cache/open_creator/skill_library/create_api/skill.json")
skill.show()
Skill Details:
• Name: create_api
• Description: Generates a CodeSkill instance using different input sources.
• Version: 1.0.0
• Usage:
create_api(request, messages=messages, messages_json_path=messages_json_path, file_content=file_content,
file_path=file_path, skill_path=skill_path, skill_json_path=skill_json_path,
huggingface_repo_id=huggingface_repo_id, huggingface_skill_path=huggingface_skill_path)
• Parameters:
• request (string): String detailing the skill functionality.
• Required: True
• messages (array): Messages as a list of dictionaries.
• messages_json_path (string): Path to a JSON file containing messages.
• file_content (string): String of file content.
• file_path (string): Path to a code/API doc file.
• skill_path (string): Directory path with skill name as stem.
• skill_json_path (string): File path with skill.json as stem.
• huggingface_repo_id (string): Identifier for a Huggingface repository.
• huggingface_skill_path (string): Path to the skill within the Huggingface repository.
• Returns:
• CodeSkill (object): The created skill.
In [ ]:
Copied!
Last update:
October 31, 2023
Created: October 18, 2023
Created: October 18, 2023