Claude Skill MartBrowse skillsQuick linesLearn by videoTerminal guideWhat is a Skill?
 
← Back to list

AI Security Assessment (ai-security)

An AI/ML-specific security skill that scans for prompt injection and jailbreaks, scores model inversion and data poisoning risk, and maps findings to MITRE ATLAS.

Security & ReviewAdvanced★ 24,151⑂ 3,405AI score 8/10Last updated: Aug 9, 2026

What it does

  • Detects prompt injection signatures (role override, system prompt extraction, tool abuse, poisoning markers) with severity and MITRE ATLAS technique IDs.
  • Scores model inversion risk by access level (black/gray/white-box) and data poisoning risk by fine-tuning scope, with mitigations for each tier.
  • Recommends guardrail patterns: input filters, output filters for PII and system-prompt leakage, and agent tool approval gates.
  • Ships three workflows (20-minute quick scan, full assessment, CI/CD deployment gate) plus seven anti-patterns to avoid.

Who it's for

  • Teams shipping LLM features who need a security review before exposing them to untrusted users.
  • Security engineers handling indirect injection risk in RAG pipelines and tool-using agents.
  • AppSec/red-team practitioners who must report AI threat coverage against MITRE ATLAS.

Example uses

  1. "Scan our chatbot before launch" → runs the scanner on seed plus domain prompts and blocks deployment when overall_risk is critical (exit code 2).
  2. "Assess poisoning risk for our fine-tuning pipeline" → risk scores per fine-tuning/RLHF/RAG scope with provenance and vetting controls.
  3. "Review an agent that can delete files" → tool-abuse vector table plus human approval gates, parameter validation and audit logging design.

· · · Install guide · · ·

Try it now, no install

Paste this into Claude to use the skill without installing anything.

Read the instructions in this file and follow them to help me:
https://raw.githubusercontent.com/alirezarezvani/claude-skills/HEAD/.gemini/skills/ai-security/SKILL.md

What I want: (describe your task here)

If Claude can't open the link, open it yourself and paste the contents instead.

↓ If it works for you, download the ZIP below and install it. Then it runs on its own — no pasting each time.

Install in the Claude app (no terminal)
  1. Download the ZIP with the button below.
  2. In Claude, open Settings → Capabilities and turn on 'Code execution and file creation'. (one time)
  3. Go to Customize → Skills → + → 'Upload a skill' and upload the ZIP.
↓ Download ZIP
Install in Claude Code

Let Claude do it — paste this into Claude Code

Install the skill I found on Claude Skill Mart.
Copy the .gemini/skills/ai-security folder from the GitHub repo alirezarezvani/claude-skills into my ~/.claude/skills/ai-security/.
When it's done, tell me in one line what this skill can do.

Install with a command instead

git clone https://github.com/alirezarezvani/claude-skills.git && mkdir -p ~/.claude/skills && cp -r claude-skills/.gemini/skills/ai-security ~/.claude/skills/ai-security

⚠ This is a third-party skill. Check the source repository before installing.

  1. Open a terminal.
  2. Clone the repository: git clone https://github.com/alirezarezvani/claude-skills.git
  3. Create the skills directory: mkdir -p ~/.claude/skills
  4. Copy the skill: cp -r claude-skills/.gemini/skills/ai-security ~/.claude/skills/ai-security
  5. Verify contents with ls ~/.claude/skills/ai-security — check for SKILL.md, scripts/ai_threat_scanner.py and references/ (without the script you still get the methodology tables).
  6. Confirm Python 3 is available: python3 --version (installing jq helps with the JSON examples).
  7. Restart Claude Code and ask something like "Run an AI security assessment on our LLM feature for prompt injection" to trigger the skill.
View source on GitHub ↗License: MIT