Scriptling Scriptling

LLM Code Generation Guide

A quick reference for Large Language Models generating Scriptling code.

Quick Summary

When generating Scriptling code:

  1. Indentation: Use 4-space indentation for blocks
  2. Booleans and Null: Use True/False for booleans, None for null (all capitalized)
  3. Loops: Use range(n), range(start, stop), or range(start, stop, step) for numeric loops
  4. Slicing: Use slice notation: list[1:3], list[:3], list[3:], list[::2], list[::-1]
  5. Dict Iteration: Use keys(dict), values(dict), items(dict) for dictionary iteration
  6. HTTP Response: HTTP functions return a Response object with status_code, body, text, headers, url fields
  7. HTTP Options: HTTP functions accept optional options dictionary with timeout and headers keys
  8. Import Libraries: Use import json, import requests, import re to load libraries
  9. JSON Functions: Always use json.loads() and json.dumps() for JSON (dot notation)
  10. HTTP Functions: Always use requests.get(), requests.post(), etc. for HTTP (dot notation)
  11. Regex: Use re.match(), re.search(), re.findall(), re.sub(), re.split() for regex
  12. Timeouts: Default HTTP timeout is 5 seconds if not specified
  13. Conditions: Use elif for multiple conditions
  14. Augmented Assignment: Use augmented assignment: x += 1, x *= 2, etc.
  15. Loop Control: Use break to exit loops, continue to skip iterations
  16. Placeholder: Use pass as a placeholder in empty blocks
  17. List Append: append(list, item) modifies list in-place
  18. String Concat: Strings use + for concatenation
  19. File Extension: Use .py file extension
  20. Status Checks: Check response.status_code before processing
  21. Error Handling: Use try/except/finally for error handling
  22. Raise Errors: Use raise "message" or raise ValueError("msg") to raise errors
  23. Unpacking: Multiple assignment: a, b = [1, 2] for unpacking lists
  24. Variadic Args: Use *args to collect extra positional arguments into a list
  25. Keyword Args: Use **kwargs to collect extra keyword arguments into a dictionary

Key Differences from Python

  • No del Statement: Use slice assignment instead — e.g. lst = lst[:n] rather than del lst[n:]
  • No Nested Classes: Classes must be defined at the top level of a module
  • Single Inheritance Only: Multiple inheritance is not supported
  • HTTP Response Format: Response object with status_code, body, text, headers, url fields
  • Default Timeout: HTTP requests have a 5-second default timeout
  • No Generators: yield is not supported
  • No Dict Comprehensions: Only list comprehensions are supported

Code Template

import json
import requests

# HTTP with headers and status check
options = {
    "timeout": 10,
    "headers": {"Authorization": "Bearer token"}
}
resp = requests.get("https://api.example.com/data", options)

if resp.status_code == 200:
    data = json.loads(resp.body)
    # Process data
    result = {"success": True, "count": len(data)}
else:
    result = {"success": False, "error": resp.status_code}

# Return structured data
scriptling.mcp.tool.return_object(result)

Common Patterns

Error Handling

try:
    result = risky_operation()
except Exception as e:
    result = None
finally:
    cleanup()

Loop with Index

for item in items(data):
    key = item[0]
    value = item[1]
    print(key, value)

String Building in Loops

# Use join() for efficiency
parts = []
for item in items:
    parts.append(str(item))
result = "".join(parts)

HTTP POST with JSON

import json
import requests

payload = {"name": "Alice", "email": "[email protected]"}
body = json.dumps(payload)
options = {"timeout": 10, "headers": {"Content-Type": "application/json"}}

response = requests.post("https://api.example.com/users", body, options)

if response.status_code == 201:
    created = json.loads(response.body)
    print("Created user:", created["id"])

See Also