A powerful context window optimizer designed to analyze, audit, and compress AI agent token utilization.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install context-engineer
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 context-engineer 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 Context Engineer?
Context Engineer is a specialized utility for developers looking to maximize the efficiency of their AI agents. It provides a deep-dive analysis of how tokens are distributed across system prompts, tool definitions, and memory files. By identifying redundancy and bloat, it ensures that your Openclaw Skills operate within specific token budgets, leading to faster response times and lower API costs.
This skill acts as an observability layer for the agent's internal state. It helps developers understand the hidden costs of installed tools and provides actionable recommendations to reclaim space in the context window. Whether you are managing large MEMORY.md files or complex system prompts, Context Engineer gives you the visibility needed to maintain a lean and effective agent environment.
Context Engineer Use Cases
- Analyzing workspace files like MEMORY.md or SOUL.md to identify and remove stale or bloated data.
- Auditing tool definitions to find overlapping functionality and excessive overhead.
- Calculating exact context window usage against a specific model budget (e.g., 128k or 200k tokens).
- Comparing snapshots before and after prompt engineering to measure real-world token savings.
- Optimizing system prompts for better compression and higher efficiency scores.
How Context Engineer Works
- The skill scans the specified Openclaw workspace directory to locate core markdown files and configuration settings.
- It performs a token estimation (approximately 4 characters per token) to calculate the footprint of static content.
- The tool audits registered tool definitions to determine the per-tool token cost and identifies unused or redundant schemas.
- It generates a comprehensive report via the terminal, highlighting efficiency scores and specific areas for optimization.
- Users can save snapshots of their current context state to perform comparisons and track improvements over time.
Context Engineer Setup
Ensure you have Python 3 installed on your system. To begin optimizing your Openclaw Skills, use the following commands:
# Run a basic analysis on your workspace
python3 skills/context-engineer/context.py analyze --workspace ~/.openclaw/workspace
# Audit tool definitions for overhead
python3 skills/context-engineer/context.py audit-tools --config ~/.openclaw/openclaw.json
# Generate a full report in the terminal
python3 skills/context-engineer/context.py report --workspace ~/.openclaw/workspace --format terminal
Context Engineer Data Schema & Taxonomy
Context Engineer organizes its analysis based on the structure of your agent's environment. The following data points are tracked:
| Component | Analysis Type | Description |
|---|---|---|
| System Prompt | Compression Audit | Detects redundancy and potential for token reduction in system instructions. |
| Tool Definitions | Overhead Calculation | Measures the token cost of each tool schema and identifies overlaps. |
| Memory Files | Bloat Detection | Scans MEMORY.md and SOUL.md for stale entries and high-volume text. |
| Snapshots | Delta Tracking | Saves JSON files representing the context state for before/after comparison. |
Context Engineer Advanced Features
- Snapshot Comparison: Directly compare two JSON snapshots to visualize projected token savings and efficiency gains.
- Custom Token Budgets: Configure the analysis to match specific model limits, ensuring your Openclaw Skills never overflow the context window.
- Efficiency Scoring: Receive a numerical grade on your context utilization based on the ratio of static content to available conversation space.
- Tool Redundancy Audit: Automatically flags tools that share similar descriptions or parameters to prevent unnecessary context consumption.
SKILL.md
Loading
Related Openclaw Skills
CacheForge
CacheForge is a token optimization gateway that streamlines onboarding and operations for OpenAI-compatible agent workflows.
vibe-check
An automated auditing tool that identifies unreviewed AI-generated code patterns and produces a scored quality report card.
CacheForge Stats
A terminal-based dashboard for monitoring CacheForge LLM usage, token savings, and performance metrics.
CacheForge Setup
A streamlined onboarding tool to register, configure, and integrate CacheForge as an LLM proxy for AI agents.
continuity-kernel
A high-reliability continuity kernel providing fail-open context injection, deterministic state persistence, and shadow-mode evaluation receipts.
Feed Diet
Analyze and optimize your Hacker News and RSS consumption with AI-generated reports and personalized digests.
METADATA
Featured*
