SkillSpector

Mentions · 7d1
Mentions · 30d2
All mentions4
First seenAug 26, 2026
Last mentioned2d ago

Timeline

4 articles · newest first

StealthyPoison poisons third-party skills to mislead LLM agents

Payloads built with one surrogate transferred unchanged to three held-out models and succeeded in a Claude Code workflow using Opus 4.8.

arXiv cs.CRNov 9, 10:05 UTC · 2d agoAI safety & security

Scanner finds local and composite vulnerabilities in public agent skills

Analysts confirmed 53 of 59 GitHub findings, and mitigations blocked attacks in 47 of 55 tested cases.

arXiv cs.CROct 27, 19:43 UTC · 15d agoAI safety & security1

Pretext: Defeating Malicious Skill Detection Frameworks for AI Agents

Pretext builds malicious agent skills that evade SkillSpector-style scanners at up to 97 percent.

arXiv cs.CRSep 30, 12:26 UTC · Sep 30, 2026AI safety & security3 · 1 read

Hottest cybersecurity open-source tools of the month: August 2026

August's open-source security roundup highlights NVIDIA SkillSpector for AI agent skill scanning, Future AGI, Chainloop, PentestGPT, and Hazmat agent containment.

Help Net SecurityAug 26, 00:00 UTC · Aug 26, 2026Tools2 · 1 read

Appears with

Claude Codeproduct · 3
NVIDIAvendor · 3
Chainloopproduct · 1
NVIDIAorganization · 1
AgentSkillsScannerproduct · 1
OpenClawproduct · 1
SkillsGuardproduct · 1
Hazmatproduct · 1
Opus 4.8ai model · 1
SkillSieveproduct · 1
PentestGPTproduct · 1
GPT-5.4-miniai model · 1
Future AGIproduct · 1
Skill Scannerproduct · 1
Ciscovendor · 1
GitHuborganization · 1

Entities are extracted by the model from each article. First seen Aug 26, 2026, last mentioned 2d ago. Watching keeps it in this browser only (no account); the watchlist page and dashboard alerts use it.