/shutdown

A pod is killed. What happens to in-flight requests, open transactions, half-written files? This skill checks shutdown implementation detail.

What this is for

• Signal handlers that call process.exit(0) immediately (drops in-flight)
• DB connection pools without drain in shutdown (queries interrupted)
• File writers without flush (data loss)
• Read-only skill. No code changes.

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 /shutdown -help or /shutdown -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

& "<SKILL_DIR>/scripts/shutdown-detail-scan.ps1" -ProjectDir "<path>"

Step 4 - Classification

Read each shutdown path:

• Safe: complete drain before exit
• Mostly-safe: drains but has timeout gaps
• Risky: drops in-flight work
• Dangerous: data corruption on termination

Step 5 - Write report

File shutdown-grade-report.md in current working directory:

1. Summary - shutdown paths by classification.
2. Path table - dangerous first. Per path: process, mechanism, grace timeout, drains requests, flushes buffers, releases connections, handles errors, grade.
3. Open questions.

Step 6 - Summarize

State report path, highlight processes that would corrupt data if terminated.

Usage

/shutdown               # interactive
/shutdown <dir>         # scan project
/shutdown -help