⚡ Optimize description processing by reducing redundant lower() calls

[frostmute]

💡 What: Optimized SkillConverter._transform_frontmatter by assigning description.lower() to a local variable desc_lower instead of calling .lower() twice in a condition.
🎯 Why: Reduces redundant string operations and potential allocations, making the conversion process more efficient, especially for skills with long descriptions.
📊 Measured Improvement: In synthetic benchmarks with 1,000,000 iterations and long description strings:

• Case where keywords are missing (worst-case): ~36% improvement (2.23s -> 1.42s).
• Case with one keyword: ~9% improvement (0.65s -> 0.59s).
• Case with both keywords: ~1.5% neutral (0.56s -> 0.57s) due to the slight overhead of assignment when the first check would have short-circuited quickly, but overall a net win for the most expensive path.

---

PR created automatically by Jules for task 11132202528257124872 started by @frostmute

[google-labs-jules]

👋 Jules, reporting for duty! I'm here to lend a hand with this pull request.

When you start a review, I'll add a 👀 emoji to each comment to let you know I've read it. I'll focus on feedback directed at me and will do my best to stay out of conversations between you and other bots or reviewers to keep the noise down.

I'll push a commit with your requested changes shortly after. Please note there might be a delay between these steps, but rest assured I'm on the job!

For more direct control, you can switch me to Reactive Mode. When this mode is on, I will only act on comments where you specifically mention me with @jules. You can find this option in the Pull Request section of your global Jules UI settings. You can always switch back!

New to Jules? Learn more at jules.google/docs.

---

For security, I will only act on instructions from the user who triggered this task.

[gemini-code-assist]

Code Review

This pull request optimizes the _transform_frontmatter function in claw2manus/converter.py by caching the lowercase version of the description string to avoid redundant calls. Feedback suggests further improving the code's robustness and efficiency by casting the description to a string to prevent potential attribute errors and slicing it to a maximum length before processing to handle large inputs more efficiently.

[gemini-code-assist]

While this optimization reduces redundant .lower() calls, there are two additional improvements to consider for robustness and efficiency:

1. Type Safety: The code assumes description is a string. However, yaml.safe_load can return other types (like dict or list) if the input YAML is malformed, which would cause an AttributeError when calling .lower(). Casting to str ensures the converter is robust against such edge cases.
2. Large Input Efficiency: Since the description is eventually truncated to MAX_DESCRIPTION_LENGTH (1024 chars), calling .lower() on a potentially massive string before truncation is inefficient. Slicing the string to the maximum allowed length before processing avoids unnecessary memory allocation and CPU cycles for invalidly long inputs.

Note: Slicing a string in Python with [:N] is zero-cost if the string is already shorter than N, as it returns the same object.

Suggested change

	desc_lower = description.lower()
	if "what it does" not in desc_lower and "when to use it" not in desc_lower:
	description = str(description)
	desc_lower = description[:ManusSkillValidator.MAX_DESCRIPTION_LENGTH].lower()
	if "what it does" not in desc_lower and "when to use it" not in desc_lower: