Lume Skill Detection
Scan any Claude or agent skill before you install it. 35 checks for prompt injection, credential theft, exfiltration, and supply-chain risk, with a 0-100 risk score and a fix for every finding.
$19
Works with the AI Agents you already use
Lume Skill Detection
Example session with this skill installed
Scan the skill in ./downloaded-skill before I install it.
- Read your context and instructions
- Compiled the lume detection
Verdict: malicious (92/100). ATT-001 instruction override in SKILL.md line 14; ATT-002 hidden system prompt; ATT-027 obfuscated payload. Fix: remove the override block and decode/review the base64 section before use.
Connects securely to your tools. The creator never sees your data.
About this skill
Lume Skill Detection inspects a Claude or other agent skill before you install it, looking for prompt injection, credential theft, covert exfiltration, unpinned dependencies, and other unsafe behavior. It runs 35 deterministic checks drawn from OWASP LLM Top 10, MITRE ATLAS, and NIST AI RMF, then scores the artifact from 0–100 with a clean, suspicious, or malicious verdict and a concrete fix for each finding. A bundled Python scanner works offline with no server, so you can review a skill folder or zip in seconds.
Includes a stdlib-only Python scanner (scripts/scan.py) that runs offline. Free open-source web version: https://github.com/ritvikindupuri/Lume---Skill-detection
Find more skills like this on the Claude Skills Marketplace, the Codex Skills Marketplace and the Gemini Skills Marketplace, or browse Claude Skills for Security.
Pattern-based checks can miss novel or heavily obfuscated attacks; treat the verdict as a review aid, not a guarantee.
How to use
- 1
Connect your agent to the Agensi MCP
Copy the prompt and paste it into Claude, ChatGPT, Cursor or any agent that supports MCP. It sets up the connection for you. Manual setup
- 2
Sign in to Agensi
Your agent shows a sign-in link. Sign in and click Allow.
- 3
Ask for what you need
Your agent loads this skill and uses it on your task. It always asks before buying a skill.
Reviews
No reviews yet
Be one of the first to try it. Every listed skill goes through the trust checks below.
Security scanned
Checked by our 8-point scan before listing
Fresh listing
Recently published to Agensi
Money-back guarantee
Doesn't do what its page promises? You get your money back.
Trust & safety
Security scan
Security scan passed 6 days ago
- Software
Git
- Optional
Python 3.8+
Permissions
Reads the target skill files only; no network access.