An AI conversation efficiency analyzer that tracks costs, session KPIs, and usage patterns across platforms like Claude and Cursor.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install memory-forge
2. Manual Installation
Copy the skill folder to one of these locations
~/.openclaw/skills/ <project>/skills/ Priority: Workspace > Local > Bundled
3. Prompt Installation
Copy this prompt to OpenClaw to install it automatically.
Help me install memory-forge using Clawhub. If Clawhub is not installed, install it first (npm i -g clawhub).
Prefer to download?
Get the raw skill files in a ZIP archive.
What is Memory Forge?
Memory Forge is a specialized analytics tool designed to provide deep insights into your interactions with Large Language Models. By analyzing local conversation history from platforms such as Claude, ChatGPT, and Cursor, it empowers developers to understand their usage patterns and optimize their expenditures. As a part of the Openclaw Skills ecosystem, this tool acts as a diagnostic layer, transforming raw session data into actionable reports.
Memory Forge focuses on privacy and efficiency, processing all data locally to ensure that sensitive conversation logs remain secure. Whether you are managing personal API costs or optimizing team-level AI workflows, it delivers the metrics needed to make data-driven decisions about how you interact with AI agents.
Memory Forge Use Cases
- Tracking API consumption costs across different AI models and projects.
- Identifying high-frequency, low-efficiency usage patterns in developer conversations.
- Analyzing project-specific engagement and topic distribution.
- Generating weekly efficiency reports to monitor AI productivity trends.
- Diagnosing expensive conversation sessions for optimization opportunities.
How Memory Forge Works
- The skill identifies conversation data stored locally in directories like ~/.claude/projects/.
- A Python-based analysis script parses JSONL conversation files to extract sessions, turns, and token usage.
- The system calculates cost breakdowns based on specific model pricing and usage duration.
- Structured statistics are generated, covering KPIs, weekly trends, and per-project distributions.
- The agent provides a diagnostic report with actionable recommendations for improving interaction efficiency.
Memory Forge Setup
To get started with this component of Openclaw Skills, ensure the scripts are accessible in your local environment. Use the following commands to initialize and run analysis:
# Run the standard statistics script with weekly trends
python3 ~/memory-forge/skill/scripts/analyze.py --weekly
# Filter analysis for a specific project
python3 ~/memory-forge/skill/scripts/analyze.py --project "my-project"
# Install the full visualization suite if needed
pip install memory-forge[all]
mforge init
mforge run
Memory Forge Data Schema & Taxonomy
Memory Forge organizes its analytics into structured formats to facilitate easy interpretation of AI usage data. The output includes:
| Data Component | Description |
|---|---|
| summary | High-level KPIs including total sessions, cost, and active days. |
| weekly | Comparative stats for the last 4-8 weeks to identify trends. |
| projects | A breakdown of costs and session counts per individual project. |
| models | Usage statistics categorized by specific LLM models (e.g., Claude 3.5 Sonnet). |
| cost_breakdown | Detailed split of expenditures across different providers. |
Memory Forge Advanced Features
- Multi-model cost tracking for granular budget management.
- Local-first processing architecture ensuring zero data leakage to external servers.
- Weekly trend reporting with automatic week-over-week change detection.
- Deep analysis hooks for topic-level insights using topics.json and extracted decision files.
- Full dashboard integration via the mforge serve command for visual usage graphs.
SKILL.md
Loading
Related Openclaw Skills
Daily Menu
An intelligent meal planning agent that transforms available ingredients into complete, step-by-step cooking menus tailored to your tastes.
Steve Krug UX Audit
Perform a comprehensive usability audit on any website or web app using Steve Krug's Don't Make Me Think methodology.
Monero Wallet Security Guide
A comprehensive security framework for protecting Monero private keys and managing XMR transactions with maximum privacy and safety.
Video Face Swap Guidance
A specialized guidance skill that streamlines the workflow for creating high-quality AI face swap videos through the VideoAny platform.
Error 404 Recovery Skill
A specialized diagnostic utility designed to help AI agents identify and rectify missing file paths or broken resource links.
GLM-OCR SDK
A specialized tool for converting images and PDFs into structured Markdown and JSON using Zhipu's high-performance cloud API.
Featured*
