Memory Processor

Process changed files and update relevant memory file sections (CLAUDE.md or AGENTS.md, whichever is the active file per project config) following official guidelines.

Guidelines

MANDATORY: All rules below must be followed exactly. Violations produce incorrect CLAUDE.md content.

@../shared/references/guidelines.md

Algorithm

1. Parse context: Read context provided by memory-updater agent:

- Changed files with categories - File content summaries - Detected dependencies - Git context (commits, diffs) - Target memory files (CLAUDE.md or AGENTS.md per project config)

1. Categorize changes: Map files to CLAUDE.md sections using the tables in "Section Names" below. Match changed files to their update triggers.

1. Analyze impact: Determine what needs updating:

- New build commands added? - Architecture changed (new dirs, renamed components)? - New coding patterns detected? - Dependencies added/removed?

1. Verify and update content: Before modifying documented content, verify accuracy:

Key distinction - conventions vs patterns: - conventions: Explicit rules humans decided (naming, imports, formatting) - patterns: Implicit patterns AI detected from recurring code structures

Removal verification: - Read the relevant CLAUDE.md section to get currently documented items - For each item that appears missing from changed files: - Use Grep to search the codebase for that item - Search in relevant directories excluding node_modules, vendor, .git - If item exists elsewhere: keep it documented - If item is not found anywhere: mark for removal

Stale command detection: - Compare documented commands against commands that actually executed successfully - If documented command differs from successful execution, update to match what worked - Examples: - Documented: python pytest | Actually worked: python -m pytest → Update - Documented: npm test | Actually worked: npm run test → Update - Documented: pytest tests/ | Actually worked: uv run pytest → Update - Source: Successful Bash tool executions from session context or git commit history

Examples: - Pattern: @decorator removed → search grep -r "@decorator" src/ - Convention: async/await style removed → search for async function or await - Architecture: utils/ directory deleted → verify no utils/ references remain - Build command: npm run dev removed from package.json → verify script is gone

1. Update memory file: Modify relevant sections in the active memory file:

- Preserve AUTO-MANAGED markers - Never touch MANUAL sections - Apply content rules (specific, concise, structured) - Do not modify CLAUDE.md redirect files (files containing only "Read AGENTS.md...")

1. Validate: Ensure updates follow guidelines:

- No generic instructions added - Specific and actionable content - Proper markdown formatting

Marker Syntax

CLAUDE.md uses HTML comment markers for selective updates:

<!-- AUTO-MANAGED: section-name -->
Content that will be automatically updated
<!-- END AUTO-MANAGED -->

<!-- MANUAL -->
Content that will never be touched
<!-- END MANUAL -->

Section Names

Root CLAUDE.md Sections

Subtree CLAUDE.md Sections

Token Efficiency

• Keep sections concise - bullet points, not paragraphs
• Use imports (@path/to/file) for detailed specs
• Follow Content Rules above (< 500 lines, stay current)

Output

Return a brief summary:

• "Updated [section names] in [memory file path] based on changes to [file names]"
• "Removed [pattern] from [section] - no longer used in codebase"
• "No updates needed - changes do not affect documented sections"