A comprehensive command-line utility for tracking, auditing, and analyzing the execution history of OpenClaw Agent tool calls.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install exec-inspector
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 exec-inspector 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 Exec Inspector?
Exec Inspector is a specialized observability tool designed for the OpenClaw ecosystem. It provides developers and power users with the ability to inspect the execution history of the exec tool, which is used by agents to run shell commands. By parsing session logs, the tool offers a transparent view of every command an agent has triggered, including timestamps, model metadata, and token usage.
As a core component for those using Openclaw Skills, it bridges the gap between agent autonomy and developer oversight. It allows for the auditing of automated tasks, debugging of command failures, and the generation of usage reports to optimize agent performance and security. Whether you are running a single session or managing complex multi-agent workflows, this skill ensures you never lose track of what your AI is doing in your terminal.
Exec Inspector Use Cases
- Auditing an agent's recent terminal activity to ensure security compliance.
- Generating statistical reports on the most frequently used CLI tools by an agent.
- Debugging session failures by searching for specific command patterns in historical logs.
- Real-time monitoring of agent actions during long-running autonomous tasks.
- Exporting command history to JSON for external data analysis or reporting.
How Exec Inspector Works
- The OpenClaw agent executes shell commands using the exec tool, which records the activity in JSONL session files.
- Exec Inspector scripts access these logs located in the session directory to extract structured data.
- The tool utilizes jq to filter and format the raw JSON data into human-readable lists or statistical summaries.
- Users can invoke the utility through shell commands or directly via AI prompts to see history, stats, or specific searches.
- For live updates, a background daemon monitors the session files and streams new command detections to a dedicated monitor terminal.
Exec Inspector Setup
To get started with this component of Openclaw Skills, follow these steps:
- Ensure you have
jqinstalled for JSON processing:
sudo apt install jq # On Ubuntu/Debian
Create the execution history script at
~/.openclaw/scripts/exec-history.shusing the provided template in the repository.Set the appropriate permissions for the scripts:
chmod +x ~/.openclaw/scripts/exec-history.sh
chmod +x ~/.openclaw/scripts/exec-monitor-daemon.sh
- (Optional) Add aliases to your
~/.zshrcor~/.bashrcfor quick access:
alias exec-list='~/.openclaw/scripts/exec-history.sh list'
alias exec-monitor='~/.openclaw/scripts/exec-monitor-daemon.sh start'
Exec Inspector Data Schema & Taxonomy
The skill organizes data by parsing the standard OpenClaw session format. Data is stored in ~/.openclaw/agents/main/sessions/*.jsonl. The key fields analyzed include:
| Field | Description |
|---|---|
| timestamp | ISO 8601 formatted time of execution |
| message.content[0].arguments.command | The specific shell command executed by the agent |
| message.model | The specific LLM (e.g., GPT-4) that generated the command |
| message.usage | Token metrics including input, output, and total tokens |
Exec Inspector Advanced Features
- Background Daemon Support: Run a persistent monitor that tails logs and alerts you to every new command in real-time.
- Interactive AI Prompts: The skill is configured so that the AI agent can autonomously run history checks and provide insights when asked.
- Deep Search Filtering: Search command history using keywords or specific time ranges using built-in grep and jq integrations.
- Session-Specific Auditing: Isolate command logs for a single unique session ID to track specific task progressions.
- Tool Usage Analytics: Aggregate data across all tools to see a breakdown of agent capabilities used over time.
SKILL.md
Loading
Related Openclaw Skills
Stock Screener & Growth Stock Analyzer
A high-performance growth stock screening and analysis tool that transforms an AI agent into a senior equity research analyst.
Startup Financial Modeling
A comprehensive toolset for building 3-5 year financial models, revenue projections, and scenario planning for early-stage startups.
Skill Lookup
A specialized tool for searching and installing modular AI agent skills from the prompts.chat registry.
Self-Improving Agent Skill
A sophisticated memory and correction framework that enables AI agents to learn from mistakes and user feedback to prevent repeating errors.
Academic Mentor
An AI-driven research advisor designed to assist graduate students and researchers through the entire lifecycle of academic projects.
Due Diligence Analyst
An intelligent AI agent that automates multi-dimensional due diligence for investment decisions and risk assessment.
Featured*
