AgentSkillsCN

Memory Optimizer

内存优化器

SKILL.md

🧠 Agent Memory Optimizer

Version: 1.0.0 | Price: $3.99 | Author: Peru 🇵🇪

Description

Analyzes an AI agent's memory files (MEMORY.md, memory/*.md), detects duplicates, stale information, missing indexes, and structural issues. Generates an optimization report with specific recommendations and can auto-fix common issues.

Features

  • •Duplicate Detection — Fuzzy matching to find repeated information across files
  • •Staleness Analysis — Identifies outdated dates, metrics, and references
  • •Structure Audit — Checks heading hierarchy, link integrity, section organization
  • •Memory Efficiency Score — 0-100 rating of memory health
  • •Auto-Fix — Can automatically deduplicate, re-index, and reorganize
  • •Detailed Reports — Markdown report with specific, actionable recommendations

Requirements

  • •python3 (3.8+)
  • •Python packages: difflib (stdlib), re (stdlib), pathlib (stdlib)
  • •No external dependencies! Uses only Python standard library.

Installation

Copy this skill folder to your workspace. No pip install needed.

bash
chmod +x analyze.py optimize.py

Usage

Analyze Memory

bash
# Analyze current workspace (auto-detects MEMORY.md and memory/ folder)
python3 analyze.py

# Analyze a specific directory
python3 analyze.py --path /path/to/workspace

# Output report to file
python3 analyze.py --output report.md

# JSON output
python3 analyze.py --json

Apply Optimizations

bash
# Preview changes (dry run — default)
python3 optimize.py

# Apply all recommended fixes
python3 optimize.py --apply

# Apply only deduplication
python3 optimize.py --apply --only dedup

# Apply only re-indexing
python3 optimize.py --apply --only reindex

# Backup before applying
python3 optimize.py --apply --backup

All Options

code
analyze.py [OPTIONS]
  --path DIR      Workspace directory to analyze (default: current dir)
  --output FILE   Save report to file
  --json          Output as JSON
  --verbose       Show detailed analysis
  --help          Show help

optimize.py [OPTIONS]
  --path DIR       Workspace directory (default: current dir)
  --apply          Apply fixes (default: dry run)
  --only TYPE      Only apply: dedup, reindex, stale, structure
  --backup         Create .bak files before modifying
  --help           Show help

Output Format

Analysis Report

markdown
# 🧠 Memory Optimization Report
Workspace: /root/.openclaw/workspace
Analyzed: 2026-02-14 03:00 UTC

## Memory Efficiency Score: 72/100

### Summary
- Files scanned: 15
- Total entries: 234
- Duplicates found: 12
- Stale entries: 8
- Missing indexes: 3
- Structure issues: 5

## 🔴 Critical Issues
1. **12 duplicate entries** across MEMORY.md and memory/2026-02-10.md
   - "GitHub token configured" appears 3 times
   - "TTS setup complete" appears 2 times

## 🟡 Warnings
1. **8 stale entries** with dates older than 30 days
   - memory/2025-12-15.md: "Current project: X" (60 days old)

## 🟢 Suggestions
1. Consider merging memory/2026-02-01.md through memory/2026-02-05.md (low activity)
2. Add table of contents to MEMORY.md (>50 entries)

## Recommended Actions
- [ ] Remove 12 duplicate entries (saves ~2.4KB)
- [ ] Archive 8 stale entries
- [ ] Add index headers to 3 files
- [ ] Fix 5 structural issues

How It Works

  1. •File Discovery — Scans for MEMORY.md, memory/*.md, and related files
  2. •Content Parsing — Extracts entries, headers, dates, metrics from markdown
  3. •Duplicate Detection — Uses SequenceMatcher for fuzzy matching (>80% similarity)
  4. •Staleness Check — Parses dates and flags entries older than configurable threshold
  5. •Structure Analysis — Validates heading hierarchy, checks for orphan sections
  6. •Scoring — Calculates efficiency score based on weighted issue counts
  7. •Report Generation — Compiles findings into actionable markdown report

Example

bash
$ python3 analyze.py --path /root/.openclaw/workspace
🧠 Agent Memory Optimizer v1.0.0
Scanning workspace: /root/.openclaw/workspace
Found 12 memory files (45.2 KB total)
Analyzing...

Memory Efficiency Score: 72/100 ⚠️

Issues found:
  🔴 Critical: 2
  🟡 Warning: 5
  🟢 Suggestion: 3

Report saved to: memory_report.md
Run `python3 optimize.py --apply` to fix issues.