TraceSentry is a cross-platform daemon and CLI that continuously monitors cache and log artifacts and helps you manage them safely. Powered by AI, it classifies files, explains their purpose, and recommends secure wiping or deletion where appropriate. The objective is to minimize unnecessary disk usage, improve system performance, and protect privacy without compromising stability.
- Search: Allows users to search for log/cache-files or specific files with a given pattern within the file system.
- Monitoring: Tracks cache and log files within the file system.
- AI Integration: Analyzes files to understand their utility and whether deletion is recommended.
- File Wiping/Deletion: Securely wipes or deletes files based on AI recommendations and user approval.
This chapter provides installation instructions for all supported platforms. TraceSentry is a platform-independent system compatible with major operating systems (Windows, Linux, macOS). However, development and testing were only conducted on the following versions, so compatibility with other versions cannot be guaranteed:
- Windows 11--10.0
- Linux 6.1.0-28
- macOS Sequoia 15.2
For all subsequent steps, the following prerequisites are required:
- Java 21+ (JRE) installation.
- A recent Maven installation (if compiling manually).
- The ZIP archive has been extracted to the desired installation directory.
The ZIP archive is generated by the root maven project. All build and setup scripts exclusively work within the ZIP archive's structure.
To use the inspection functionality, create an OpenAI API key by following the instructions at OpenAI API Keys. Add credits to your account to enable the OpenAI API: OpenAI Billing Overview. This API key will be needed in subsequent steps. If this step is skipped, you will not have access to the inspection functionality.
To simplify installation, scripts are provided for all supported platforms in the installation directory:
build: Compiles the project, creates the executable JAR files, and replaces them in thebinfolder.setup: Sets up necessary environment variables and configures the daemon for autostart.build-setup: Sequentially executesbuildandsetup.
Since the JAR files are already included in the installation directory, manual compilation is not necessary. It is recommended to execute setup directly. For the scripts to recognize the OpenAI API key, it must be stored in a file named apikey.txt in the installation directory.
The following scripts are available for Linux:
buildsetupbuild-setup
The following scripts are available for Windows:
build.ps1setup-win.ps1build-setup-win.ps1
Before executing the scripts, ensure they can run. Warning: This action allows all scripts to execute and must be performed by an administrator. Use the following command in PowerShell:
set-executionpolicy unrestrictedRecommended execution of setup-win.ps1:
.\setup-win.ps1The following scripts are available for macOS:
buildsetup-macbuild-setup-mac
Warning: For simplicity, configurations are made exclusively in the ~/.zprofile file. These can be manually removed if necessary. The setup script should only be executed once to avoid duplicate entries in ~/.zprofile.
Ensure that the CLI and scripts can be executed using the following command:
chmod +x ts
chmod +x <script-name>If a quarantine tag is set, it can be removed with the following command:
xattr -d com.apple.quarantine <script-name>Recommended execution of setup-mac:
./setup-mac- OpenAI GPT-4o mini
- Java/Spring/Maven
- SQLite/H2
- Create a new branch, prefixed with
feat/(git checkout -b feat/feature-name) - Commit your changes (
git commit -m 'Add feature') - Push to the branch (
git push origin feat/feature-name) - Create a merge request
This project is licensed under the MIT License: LICENSE
- Group:
- Advisor: Dr. Simon Kramer