Share Usage

Share Skill Usage Statistics

This skill helps you generate and share anonymized statistics about your skill usage with the community. This helps maintainers understand which skills are most valuable and guides future development.

When to Use

Use this skill when you want to:

• Share your experience with the skill repository
• Help the community understand which skills are most useful
• Contribute usage data to improve skill development priorities
• Celebrate your research productivity with the community

What Information Will Be Shared

Before generating the report, you should understand exactly what will be shared:

✅ Information That WILL Be Shared

1. Skill usage counts - How many times you used each skill
2. Skill names - Which skills from this repository you used
3. Time period - The date range of your usage data
4. General statistics - Total skill calls, unique skills used
5. Optional context - Your research domain (if you choose to provide it)

❌ Information That Will NOT Be Shared

1. Your identity - No usernames, emails, or identifying information
2. Project details - No file paths, code, or project names
3. Conversation content - No actual research questions or discussions
4. Timestamps - Only aggregated date ranges, not specific times
5. System information - No machine names, IPs, or system details
6. Other tools - Only skills from this repository are counted

Privacy Notice

This skill generates a completely anonymized report. The report contains ONLY:

• Skill names and usage counts
• Date range (e.g., "Last 30 days")
• Optional research domain tag (e.g., "fMRI analysis", "EEG research")

No personally identifiable information, project details, or conversation content is included.

How to Use

When the user invokes this skill (e.g., "share my skill usage" or "generate usage report"), follow these steps:

Step 1: Confirm Privacy Understanding

First, present the privacy information to the user and confirm they understand:

Say to the user:

"I'll generate an anonymized report of your skill usage from this repository. The report will include:

✅ What WILL be included:

• Skill names and usage counts
• Date range (e.g., "2024-01-01 to 2024-03-01")
• Optional research domain tag (if you provide it)
• Optional comments (if you provide them)

❌ What will NOT be included:

• Your identity (no usernames, emails, names)
• Project details (no file paths, code, project names)
• Conversation content (no research questions or discussions)
• Specific timestamps (only date ranges)
• System information (no machine names, IPs)
• Skills from other repositories

The report is completely anonymized and sharing is optional. Shall I proceed?"

Wait for user confirmation before proceeding.

Step 2: Analyze Claude Code Logs

Implementation Steps

When this skill is invoked, follow these steps:

Step 1: Analyze Claude Code Logs

Create a Python script inline to analyze logs:

import json
import re
from pathlib import Path
from collections import Counter
from datetime import datetime

def analyze_usage(claude_dir="~/.claude", days=None):
    """Extract skill usage from Claude Code logs."""
    claude_dir = Path(claude_dir).expanduser()

    all_skills = []
    all_timestamps = []

    # Parse history.jsonl
    history_path = claude_dir / "history.jsonl"
    if history_path.exists():
        with open(history_path, 'r', encoding='utf-8', errors='ignore') as f:
            for line in f:
                if '"name":"Skill"' in line or '"name": "Skill"' in line:
                    match = re.search(r'"skill":\s*"([^"]+)"', line)
                    if match:
                        skill_name = match.group(1)
                        # Only count skills from this repository
                        if skill_name.startswith('awesome-cognitive-and-neuroscience-skills:'):
                            clean_name = skill_name.replace('awesome-cognitive-and-neuroscience-skills:', '')
                            all_skills.append(clean_name)

                # Extract timestamp
                timestamp_match = re.search(r'"timestamp":(\d+)', line)
                if timestamp_match:
                    ts = int(timestamp_match.group(1)) / 1000
                    all_timestamps.append(datetime.fromtimestamp(ts))

    # Parse debug files
    debug_dir = claude_dir / "debug"
    if debug_dir.exists():
        for debug_file in debug_dir.glob("*.txt"):
            if days:
                file_age = (datetime.now().timestamp() - debug_file.stat().st_mtime) / 86400
                if file_age > days:
                    continue

            with open(debug_file, 'r', encoding='utf-8', errors='ignore') as f:
                content = f.read()
                matches = re.finditer(r'"name":\s*"Skill".*?"skill":\s*"([^"]+)"', content, re.DOTALL)
                for match in matches:
                    skill_name = match.group(1)
                    if skill_name.startswith('awesome-cognitive-and-neuroscience-skills:'):
                        clean_name = skill_name.replace('awesome-cognitive-and-neuroscience-skills:', '')
                        all_skills.append(clean_name)

    if not all_skills:
        return None

    # Calculate statistics
    skill_counts = Counter(all_skills)

    if all_timestamps:
        min_date = min(all_timestamps)
        max_date = max(all_timestamps)
        date_range = f"{min_date.strftime('%Y-%m-%d')} to {max_date.strftime('%Y-%m-%d')}"
        days_span = (max_date - min_date).days
    else:
        date_range = "Unknown"
        days_span = None

    return {
        'skills': skill_counts,
        'total_calls': len(all_skills),
        'unique_skills': len(skill_counts),
        'date_range': date_range,
        'days_span': days_span,
        'most_used': skill_counts.most_common(1)[0] if skill_counts else None
    }

Step 3: Ask User for Optional Information

Use the AskUserQuestion tool to gather optional information:

Question 1: Research Domain

• Header: "Domain"
• Question: "What research domain do you work in? (This helps others find relevant skills)"
• Options:
  • "fMRI analysis"
  • "EEG/MEG research"
  • "Cognitive modeling"
  • "Behavioral experiments"
  • "Other (specify in notes)"

Question 2: Comments

• Header: "Experience"
• Question: "Would you like to add comments about your experience with these skills?"
• Options:
  • "Yes, I'll add comments"
  • "No comments"

Question 3: Sharing Preference

• Header: "Action"
• Question: "What would you like to do with the report?"
• Options:
  • "Share to GitHub Discussions (Recommended)"
  • "Save locally only"
  • "Just preview, don't save"

If user chose "Yes" for comments, ask them to provide their comments in a follow-up message.