SkillSpector by NVIDIA is an open-source AI skill security scanner that checks Claude Code, Cursor and agent skills for vulnerabilities BEFORE you install them. In this honest overview I explain what SkillSpector is, how it works, and run it on 3 real skills — every one scored 100/100 CRITICAL.
----
???? DYNAMOUS AI COMMUNITY
Want to learn agentic coding with live daily events and workshops?
Check out Dynamous AI: https://dynamous.ai/?code=646a60
Get 10% off here ???? https://shorturl.smartcode.diy/dynamous_ai_10_percent_discount
⚡ HOSTINGER — RELIABLE HOSTING FOR YOUR PROJECTS (10% OFF)
Whether you're shipping a portfolio, a side project, n8n flows, or AI agents — I use Hostinger for fast, affordable VPS + web hosting.
Get 10% off here ???? https://hostinger.com/DIYSMARTCODE
(Affiliate link — costs you nothing, supports the channel.)
----
What you will see in this 2-minute breakdown:
→ Why SkillSpector exists — 26% of AI skills carry vulnerabilities, 5% look malicious
→ What it is — open-source NVIDIA scanner that reads every file before you install a skill
→ How it works — 64 attack patterns across 16 categories (prompt injection, data exfiltration, privilege escalation, supply chain, excessive agency, code execution)
→ The command — skillspector scan on a folder, repo, zip, or URL, with terminal / markdown / JSON reports
→ 3 real scans — archon (19 findings, 4 HIGH), agent-browser (ships shell scripts), and a marketplace brainstorming skill — all 100/100
→ The honest caveat — static scan, patterns in docs and helper scripts, signal not final verdict
SkillSpector (NVIDIA): https://github.com/NVIDIA/skillspector
Three skills — including one from the official Claude plugin marketplace — all scored a perfect 100/100 CRITICAL from static patterns alone. Real red flag you'd act on, or is the scanner over-flagging documentation examples? Would YOU still install a skill that scored 100? Drop your honest verdict below.
#nvidia #skillspector #aisecurity #aiagents #claudecode #claudeai #cursor #aitools #devtools #cybersecurity #promptinjection #supplychainsecurity #opensource #vulnerabilityscanner #agenticai #aicoding #infosec #appsec #mcp #aisafety
----
???? DYNAMOUS AI COMMUNITY
Want to learn agentic coding with live daily events and workshops?
Check out Dynamous AI: https://dynamous.ai/?code=646a60
Get 10% off here ???? https://shorturl.smartcode.diy/dynamous_ai_10_percent_discount
⚡ HOSTINGER — RELIABLE HOSTING FOR YOUR PROJECTS (10% OFF)
Whether you're shipping a portfolio, a side project, n8n flows, or AI agents — I use Hostinger for fast, affordable VPS + web hosting.
Get 10% off here ???? https://hostinger.com/DIYSMARTCODE
(Affiliate link — costs you nothing, supports the channel.)
----
What you will see in this 2-minute breakdown:
→ Why SkillSpector exists — 26% of AI skills carry vulnerabilities, 5% look malicious
→ What it is — open-source NVIDIA scanner that reads every file before you install a skill
→ How it works — 64 attack patterns across 16 categories (prompt injection, data exfiltration, privilege escalation, supply chain, excessive agency, code execution)
→ The command — skillspector scan on a folder, repo, zip, or URL, with terminal / markdown / JSON reports
→ 3 real scans — archon (19 findings, 4 HIGH), agent-browser (ships shell scripts), and a marketplace brainstorming skill — all 100/100
→ The honest caveat — static scan, patterns in docs and helper scripts, signal not final verdict
SkillSpector (NVIDIA): https://github.com/NVIDIA/skillspector
Three skills — including one from the official Claude plugin marketplace — all scored a perfect 100/100 CRITICAL from static patterns alone. Real red flag you'd act on, or is the scanner over-flagging documentation examples? Would YOU still install a skill that scored 100? Drop your honest verdict below.
#nvidia #skillspector #aisecurity #aiagents #claudecode #claudeai #cursor #aitools #devtools #cybersecurity #promptinjection #supplychainsecurity #opensource #vulnerabilityscanner #agenticai #aicoding #infosec #appsec #mcp #aisafety
- Category
- Cybersecurity
- Tags
- ai, ai coding, artificial intelligence

Comments