Claude Skills by msiShariful

Quick one-liner health check of your context window. Shows usage percentage, fill level, and whether bloat is detected — without interrupting your flow. Use when you want a fast pulse check before starting a task.

Find large context items that are not relevant to the current request. Identifies files, documents, and instructions that could be removed to free up token budget. Use when context is running out or responses are being truncated.

Analyze conversation history turn-by-turn and identify the exact point to run /clear from to prune stale history while keeping everything relevant to the current task. Use when history has grown long but you want to continue in the same session.

Analyze the current context window and report a breakdown of what's consuming tokens — system prompts, files, conversation history, skills, and MCP configs. Use when the user wants to understand their token budget.

Audit all connected MCP servers and their tool definitions. Shows which servers are loaded, how many tokens each costs, and which tools haven't been used — so you can disconnect what you don't need. Use when you have many MCP servers connected and want to reclaim token budget.

Analyze CLAUDE.md and other loaded instruction files for verbosity, redundancy, and bloat. Produces a condensed rewrite that preserves all meaning while cutting token cost. Use when your system prompt or CLAUDE.md feels too long.

Detect when a conversation has drifted across multiple unrelated topics and recommend whether to start a new conversation. Identifies the optimal split point and summarizes what to carry forward. Use when a session has been running long or has covered many different tasks.

Estimate how many tokens a file, directory, URL, or pasted content would add to your context before you load it. Use before opening large files or reading external content to avoid blowing your context budget.