--- name: startup-profile-analyzer description: "Startup profile analyzer: trace the entire initialization chain, LLM judges each step as essential/lazy-loadable/suspicious. Read-only. Trigger: /startup" trigger: /startup --- # /startup What happens when your service starts? This skill traces every init step and finds cold-start optimizations. ## What this is for - Module-level imports that could be lazy-loaded - Connection pools, cache warming, scheduler registration at boot - Heavy dependencies loaded at startup but used only in one handler - **Read-only skill.** No code modification. ## PROTECTION RULE - never ~/.claude/ Read-only skill. Guard required if write mode added later. ## What You Must Do When Invoked During analysis, assign a confidence level to each finding: proven (confirmed by evidence), likely (strong signal, needs review), or suspected (weak signal). If `/startup -help` or `/startup -h` (without further arguments) is invoked: output the `## Usage` section unchanged and stop. Otherwise follow these steps in order, skipping none. ### Step 1 - Help check If invoked with `-help` or `-h`, output the `## Usage` section unchanged and stop. ### Step 2 - Clarify target Clarify `-ProjectDir`. Get confirmation. ### Step 3 - Scan ```powershell & "/scripts/startup-trace.ps1" -ProjectDir "" ``` ### Step 4 - Analysis Read each init step in context: - Is this essential at startup or could it be lazy-loaded? - What would break if deferred? - Is this a side effect at import time (suspicious)? ### Step 5 - Write report File `startup-report.md` in current working directory: 1. **Summary** - total init steps, deferrable count, estimated boot time reduction. 2. **Init chain** - sorted by phase. Per step: phase, description, classification, deferral recommendation. 3. **Suspicious findings** (side effects at import time). 4. **Open questions**. ### Step 6 - Summarize State report path, highlight top 3 deferral candidates. ## Usage ``` /startup # interactive /startup # scan project /startup -help ```