# AI use cases for AI Security Tool (ZeroDayEvil/ai-security-tool) Repo: [ZeroDayEvil/ai-security-tool](https://github.com/ZeroDayEvil/ai-security-tool) | Use cases: 4 | Page: https://dedupedai.com/use-cases/zerodayevil-ai-security-tool/ AI Security Tool is an open-source, cross-platform terminal and vulnerability scanner that combines multi-protocol remote access, CVE and SBOM supply chain auditing, and autonomous AI agents. This file holds every use case of this repo. Each one starts with a "## " heading and ends with its URL. - Scan a project's dependencies for CVEs and SBOM gaps: https://dedupedai.com/use-cases/zerodayevil-ai-security-tool/scan-project-dependencies-for-cves-and-sbom/ - Run one command across several SSH sessions at once: https://dedupedai.com/use-cases/zerodayevil-ai-security-tool/manage-multiple-remote-servers-with-broadcast-commands/ - Host the Web UI for a team with login enabled: https://dedupedai.com/use-cases/zerodayevil-ai-security-tool/host-web-interface-securely-for-a-team/ - Automate CTF challenge steps with AI agents: https://dedupedai.com/use-cases/zerodayevil-ai-security-tool/automate-ctf-challenge-steps-with-ai-agents/ ## How to scan a project's dependencies for CVEs and SBOM gaps with AI Security Tool Repo: [ZeroDayEvil/ai-security-tool](https://github.com/ZeroDayEvil/ai-security-tool) | Difficulty: Intermediate | Time: 30 minutes | Basis: From the README Generate an SBOM for a project and correlate its components with CVE data to get exportable findings. **Who it is for:** A DevSecOps engineer or application security reviewer who must check a project's third-party components before a release. **The problem:** Dependency and CVE checks are spread across separate tools, and the results are hard to tie back to a specific SBOM. **The outcome:** A console, JSON, CSV, HTML, or PDF report listing vulnerabilities enriched with remediation context for the target project. ### Steps 1. Install the Linux or macOS build with `curl -o- https://raw.githubusercontent.com/ZeroDayEvil/ai-security-tool/main/scripts/one-line-web.sh | bash`, or download the matching build from the Downloads table. 2. Start the Web UI with `./build/bin/run-prod.sh` and open http://127.0.0.1:5577 in your browser. 3. In the Web UI, open the scanning workflow and point it at your project folder or asset. (inferred, check the README) 4. Let the Binary and Dependency Analyzers generate or parse the SBOM and pass it to the Vulnerability Correlator. 5. Export the findings as JSON, CSV, HTML, or PDF and review the remediation data for each item. ### Prompt to paste into your agent ```text Using the AI Security Tool repo at https://github.com/ZeroDayEvil/ai-security-tool, run its SBOM and CVE scanning workflow against and export a JSON report. Expected output: a JSON file listing each component with its CVE matches and remediation context. Verify that the SBOM component count in the report matches the dependency count in the project's lock or manifest file. ``` ### Caveat The README does not document the exact steps to select a scan target in the Web UI, so confirm that menu path in the running build before automating it. Risk, reverse engineering: Binary analysis steps can expose internals of third-party software, so only run them on code and binaries you are authorized to analyze. Disclaimer: Third-party repos we don't own or vouch for. AI-generated use cases can be wrong: review everything and run at your own risk. Full text: https://dedupedai.com/disclaimer/ URL: https://dedupedai.com/use-cases/zerodayevil-ai-security-tool/scan-project-dependencies-for-cves-and-sbom/ ## How to run one command across several SSH sessions at once with AI Security Tool Repo: [ZeroDayEvil/ai-security-tool](https://github.com/ZeroDayEvil/ai-security-tool) | Difficulty: Intermediate | Time: 30 minutes | Basis: From the README Open SSH, SFTP, and other remote sessions in one terminal and send commands to all of them together. **Who it is for:** A sysadmin or network operator who maintains a small fleet of Linux servers and repeats the same checks on each. **The problem:** Switching between separate terminal clients for each server is slow and makes it easy to run a command on the wrong host. **The outcome:** One terminal workspace with several active remote sessions, where a single input is sent to every session you select. ### Steps 1. Install the build for your platform using the Quick Start table or the one-line script for Linux or macOS. 2. Start the app with `./build/bin/run-prod.sh` and open http://127.0.0.1:5577. 3. Add a bookmark for each server over SSH, using key or password authentication. 4. Open a session to each server, then turn on multi-input broadcast so commands go to all active sessions. 5. Run a read-only check such as a uptime or disk usage command and confirm the output appears for every host. ### Prompt to paste into your agent ```text Using the AI Security Tool repo at https://github.com/ZeroDayEvil/ai-security-tool, open SSH sessions to the hosts in and enable multi-input broadcast to run a read-only disk usage check. Expected output: one result block per host. Verify that every session returned output and that no session shows a connection error. ``` ### Caveat Broadcast sends the same input to every session at once, so use read-only commands until you have confirmed each host is the one you intended. Risk, credentials: SSH keys and passwords stored in bookmarks or synced through GitHub Gist or Gitee Snippets must be protected and rotated if exposed. Disclaimer: Third-party repos we don't own or vouch for. AI-generated use cases can be wrong: review everything and run at your own risk. Full text: https://dedupedai.com/disclaimer/ URL: https://dedupedai.com/use-cases/zerodayevil-ai-security-tool/manage-multiple-remote-servers-with-broadcast-commands/ ## How to host the Web UI for a team with login enabled with AI Security Tool Repo: [ZeroDayEvil/ai-security-tool](https://github.com/ZeroDayEvil/ai-security-tool) | Difficulty: Advanced | Time: 1 hour | Basis: From the README Deploy the web interface on a shared server with password authentication and local terminal access turned off. **Who it is for:** A team lead or platform engineer who wants several colleagues to reach the tool from their browsers on a shared server. **The problem:** A web-based security terminal exposed to a network without authentication or with local shell access enabled is a serious risk. **The outcome:** A running Web UI behind an Nginx reverse proxy, with password authentication on and local server terminal access disabled. ### Steps 1. Create a `.env` file in the project folder and set `ENABLE_AUTH=1`, `DISABLE_LOCAL_TERMINAL=1`, `SERVER_SECRET`, and `SERVER_PASS` to strong values. 2. Start the service with `./run-ai-security-tool-web.sh`. 3. Adapt the example configuration in `examples/nginx.conf` to your domain and add your SSL certificate paths. (inferred, check the README) 4. Reload Nginx and open your public domain in a browser to confirm the login page appears before the Web UI loads. 5. Confirm that the local terminal option is unavailable in the Web UI after login. ### Prompt to paste into your agent ```text Using the AI Security Tool repo at https://github.com/ZeroDayEvil/ai-security-tool, configure the .env file with ENABLE_AUTH=1 and DISABLE_LOCAL_TERMINAL=1 and start the web service for . Expected output: the service starts and the public URL shows a password prompt. Verify that an unauthenticated request to the URL is refused before the Web UI loads. ``` ### Caveat The README says the AI agents can execute system commands, so keep the host's permissions minimal and watch agent activity on the real-time dashboard. Risk, credentials: The SERVER_SECRET and SERVER_PASS values in .env are the only barrier to the Web UI, so store them safely and never commit the file to version control. Risk, remote script: The one-line installers pipe remote scripts into bash or run a downloaded batch file, so review the script before running it on a production server. Disclaimer: Third-party repos we don't own or vouch for. AI-generated use cases can be wrong: review everything and run at your own risk. Full text: https://dedupedai.com/disclaimer/ URL: https://dedupedai.com/use-cases/zerodayevil-ai-security-tool/host-web-interface-securely-for-a-team/ ## How to automate CTF challenge steps with AI agents with AI Security Tool Repo: [ZeroDayEvil/ai-security-tool](https://github.com/ZeroDayEvil/ai-security-tool) | Difficulty: Intermediate | Time: 45 minutes | Basis: From the README Use the CTF and recon agents with the LLM assistant to suggest commands and analyze terminal output during a challenge. **Who it is for:** A CTF player or security student working on a timed challenge who wants help choosing tools and reading their output. **The problem:** Choosing the next tool, tuning its parameters, and reading long terminal output takes time away from solving the challenge. **The outcome:** Suggested commands, script drafts, and summarized terminal output for each challenge step, with the human still approving each action. ### Steps 1. Install the build for your platform and start the Web UI with `./build/bin/run-prod.sh`. 2. Open http://127.0.0.1:5577 and add your DeepSeek, OpenAI, or custom API key in the LLM settings. 3. Start the CTFWorkflowManager agent and give it the challenge description and target address. 4. Review each command the agent proposes on the real-time dashboard before allowing it to run. 5. Ask the LLM assistant to analyze a terminal output block and note the findings you want to keep. ### Prompt to paste into your agent ```text Using the AI Security Tool repo at https://github.com/ZeroDayEvil/ai-security-tool, start the CTFWorkflowManager agent on the challenge at and have it propose the next three commands without running them. Expected output: a list of three commands with a one-line reason for each. Verify that no command ran before a human approved it. ``` ### Caveat The README warns that agents can execute system commands, so keep the human approval step on and do not point the agent at systems outside the challenge scope. Risk, reverse engineering: CTF exploitation techniques can be reused against real systems, so only run them against challenge environments you are allowed to test. Disclaimer: Third-party repos we don't own or vouch for. AI-generated use cases can be wrong: review everything and run at your own risk. Full text: https://dedupedai.com/disclaimer/ URL: https://dedupedai.com/use-cases/zerodayevil-ai-security-tool/automate-ctf-challenge-steps-with-ai-agents/