audit-performance

Performance Audit

Perform a comprehensive performance audit focusing on database queries, memory usage, caching, and algorithm efficiency.

Input Parsing

Parse $ARGUMENTS to extract path, level, and optional meta-instructions:

Format: <path> [level] [-- <meta-instructions>]

Arguments:
- path: Target directory or file (required, default: current directory)
- level: Audit depth - quick|standard|deep (optional, default: standard)
- -- meta-instructions: Additional focus areas or filters (optional)

Examples:
- /acc:audit-performance ./src
- /acc:audit-performance ./src deep
- /acc:audit-performance ./src quick
- /acc:audit-performance ./src -- focus on memory and caching
- /acc:audit-performance ./src deep -- focus on N+1 queries
- /acc:audit-performance ./src -- level:deep (backward compatible)

Parsing rules:

1. Split $ARGUMENTS by -- (space-dash-dash-space)
2. First part = positional arguments, Second part = meta-instructions
3. In positional arguments, check if last word is a valid level (quick|standard|deep)
4. If level found → extract it; remaining = path
5. Also accept level:quick|standard|deep in meta-instructions (backward compatibility)
6. Priority: positional > meta-instruction > default (standard)
7. Default path: current directory (if empty)

Target

• Path: First part of $ARGUMENTS (before --)
• Meta-instructions: Second part (after --) — customize audit focus

If meta-instructions provided, adjust audit to:

• Focus on specific performance categories
• Skip categories if requested
• Apply additional depth to specific areas
• Modify output verbosity

Pre-flight Check

1. Verify the path exists:

   • If $ARGUMENTS is empty, ask user for the project path
   • If path doesn't exist, report error and stop

2. Verify it's a PHP project:

   • Check for composer.json or *.php files
   • If not a PHP project, report and stop

Instructions

Use three specialized reviewers in parallel:

Task tool with subagent_type="acc:performance-reviewer"
prompt: "Perform performance audit on [PATH]. Audit level: [LEVEL]. [META-INSTRUCTIONS if provided]

Analyze for:
1. N+1 Query Problems — queries inside loops, missing eager loading
2. Query Efficiency — SELECT *, missing indexes, full table scans
3. Caching Strategy — missing cache opportunities, repeated expensive operations
4. Unnecessary Loops — nested loop inefficiency, redundant iterations
5. Lazy Loading Problems — premature loading, missing pagination
6. Batch Processing — single-item vs bulk operations, transaction overhead
7. Algorithm Complexity — O(n²) patterns, exponential growth
8. Database Index Usage — missing indexes, incorrect composite order

Provide severity classification, impact estimates, current vs optimal complexity, code examples."

Task tool with subagent_type="acc:resource-reviewer"
prompt: "Perform resource usage audit on [PATH]. Audit level: [LEVEL]. [META-INSTRUCTIONS if provided]

Analyze for:
1. Memory Issues — large arrays, missing generators, unbounded loading
2. Connection Pool Issues — leaks, connections in loops, missing cleanup
3. Serialization Overhead — large objects, N+1 during serialization
4. Async Patterns — blocking operations in request cycle
5. File I/O Patterns — full file reads, missing streaming

Provide severity classification, impact estimates, code examples."

Task tool with subagent_type="acc:scalability-reviewer"
prompt: "Perform scalability audit on [PATH]. Audit level: [LEVEL]. [META-INSTRUCTIONS if provided]

Analyze for:
1. Scalability Readiness — stateless design, session handling, hardcoded config
2. Database Scaling — read/write separation, connection pooling, replica usage

Provide severity classification, scaling impact, code examples."