vastai-host-mcp: Skill for Claude Code

Context-Aware Operations Skill

This skill provides patterns and techniques for managing large files and command outputs efficiently, preventing context window exhaustion while maintaining effective problem-solving capabilities.

When to Use This Skill

• Before reading any file in the codebase
• Before executing commands that might produce large output
• When working with logs, build outputs, or data files
• When searching through codebases
• When debugging issues that might involve large resources

Core Principle

Always check before you dump!

Never blindly dump large resources into your context. Always:

1. Check the size first
2. Use filtering if needed
3. Focus on relevant portions only

File Size Checking

Check Line Count

# Fast line count
wc -l filename.txt

# Line count without filename
wc -l < filename.txt

# Line count with human-readable file size
wc -l filename.txt && ls -lh filename.txt

Check File Size

# Human-readable size
ls -lh filename.txt

# Size in bytes (Linux)
stat -c%s filename.txt

# Size in bytes (macOS)
stat -f%z filename.txt

# Quick check if file is large
[ $(wc -l < filename.txt) -gt 500 ] && echo "Large file" || echo "Small file"

Filtered File Reading

Read Beginning and End

# First 50 lines
head -n 50 filename.txt

# Last 50 lines
tail -n 50 filename.txt

# Both ends with separator
head -n 30 filename.txt && echo "..." && tail -n 30 filename.txt

Read Specific Ranges

# Lines 100-200
sed -n '100,200p' filename.txt

# Around a specific line (line 150 ± 25 lines)
sed -n '125,175p' filename.txt

# Skip first N lines, show next M
tail -n +100 filename.txt | head -n 50

Search-Based Reading

# Find and show context
grep -n "pattern" filename.txt

# Show matching lines with 5 lines of context
grep -C 5 "pattern" filename.txt

# Show matching lines with line numbers
grep -n "pattern" filename.txt | head -20

# Count matches without showing content
grep -c "pattern" filename.txt