configure-memory-profiling（配置内存分析）

/configure:memory-profiling

Check and configure memory profiling infrastructure for Python projects using pytest-memray.

When to Use This Skill

Context

• Project root:!pwd
• Python project:!find. -maxdepth 1 \(-name 'pyproject.toml' -o -name 'setup.py' \)
• pytest-memray installed:!grep -r 'pytest-memray' pyproject.toml requirements*.txt
• memray installed:!grep -r 'memray' pyproject.toml requirements*.txt
• Conftest fixtures:!grep -l 'memray' tests/conftest.py
• Memory test files:!find tests -maxdepth 2 -name '*memory*' -o -name '*memray*'
• Benchmark tests:!find tests -maxdepth 2 -type d -name 'benchmarks'
• CI workflows:!find.github/workflows -maxdepth 1 -name '*memory*'
• Memory reports dir:!find. -maxdepth 1 -type d -name 'memory-reports'

Parameters

Parse from $ARGUMENTS:

• --check-only: Report memory profiling compliance status without modifications
• --fix: Apply all fixes automatically without prompting
• --threshold <mb>: Set default memory threshold in MB (default: 100)
• --native: Enable native stack tracking for C extensions

Supported tools:

Execution

Execute this memory profiling configuration check:

Step 1: Verify this is a Python project

Read the context values. If no pyproject.toml or setup.py is found, report "Not a Python project" and stop.

Step 2: Check latest tool versions

Use WebSearch or WebFetch to verify current versions:

1. pytest-memray: Check PyPI
2. memray: Check PyPI

Step 3: Analyze current memory profiling setup

Check for complete setup:

• pytest-memray installed as dev dependency
• memray backend installed
• pytest configuration in pyproject.toml (markers, addopts)
• Memory limit tests using @pytest.mark.limit_memory
• Leak detection enabled (--memray-leak-detection)
• Native tracking configured (if --native flag)
• CI/CD integration configured
• Reports directory exists

Step 4: Generate compliance report

Print a compliance report covering:

• Installation status (pytest-memray, memray, pytest versions)
• Configuration (pytest integration, markers, leak detection, native tracking)
• Test coverage (memory limit tests, allocation benchmarks)
• CI/CD integration (workflow, threshold, artifact upload, trend tracking)

End with overall issue count and recommendations.

If --check-only is set, stop here.

Step 5: Install and configure pytest-memray (if --fix or user confirms)

1. Install pytest-memray: uv add --group dev pytest-memray
2. Install native support if --native: uv add --group dev pytest-memray[native]
3. Update pyproject.toml with pytest configuration (markers, filterwarnings)
4. Create memory-reports/ directory
5. Use configuration templates from REFERENCE.md

Step 6: Create memory profiling test files

1. Add memory fixtures to tests/conftest.py (reports dir setup, threshold fixture, data generator)
2. Create tests/test_memory_example.py with example memory limit tests
3. Create tests/benchmarks/test_memory_benchmarks.py for trend tracking
4. Use test templates from REFERENCE.md

Step 7: Add package scripts

Add memory profiling commands to Makefile or pyproject.toml:

• test-memory: uv run pytest --memray
• test-memory-report: Run with bin output + generate flame graph
• test-memory-leaks: uv run pytest --memray --memray-leak-detection
• test-memory-native: uv run pytest --memray --native

Step 8: Configure CI/CD integration

Create .github/workflows/memory-profiling.yml with:

• Memory profiling on PRs (detect regressions)
• Scheduled weekly benchmarks for trend tracking
• Flame graph generation
• PR comment with results
• Use workflow template from REFERENCE.md

Step 9: Update standards tracking

Update .project-standards.yaml:

components:
  memory_profiling: "2025.1"
  memory_profiling_tool: "pytest-memray"
  memory_profiling_threshold_mb: 100
  memory_profiling_leak_detection: true
  memory_profiling_ci: true
  memory_profiling_native: false

Step 10: Print final compliance report

Print a summary of packages installed, configuration applied, test files created, commands available, CI/CD configured, and next steps for the user.

For detailed test templates, CI workflows, and standalone memray commands, see REFERENCE.md.

Agentic Optimizations

Flags

Examples

# Check compliance and offer fixes
/configure:memory-profiling

# Check only, no modifications
/configure:memory-profiling --check-only

# Auto-fix with custom threshold
/configure:memory-profiling --fix --threshold 200

# Enable native tracking for C extensions
/configure:memory-profiling --fix --native

Error Handling

• Not a Python project: Skip with message, suggest manual setup
• pytest not installed: Offer to install pytest first
• memray not supported: Note platform limitations (Linux/macOS only)
• Native tracking unavailable: Warn about missing debug symbols
• CI workflow exists: Offer to update or skip

See Also

• /configure:tests - Configure testing frameworks
• /configure:coverage - Code coverage configuration
• /configure:load-tests - Load and performance testing
• /configure:all - Run all compliance checks
• pytest-memray docs: https://pytest-memray.readthedocs.io
• memray docs: https://bloomberg.github.io/memray