https://skillscan.io/research/2026-10-23T02:58:32Zhttps://skillscan.io/research/2026/top-100-refresh/2026-10-22T00:00:00Z

We re-pulled the skills.sh top 100 by installs eight months after the first snapshot. Only six skills overlapped. We scanned the 476 skills behind the other 94 and found one vendor's skill set republished under at least seven owners, two false positives in the top 60, and a Microsoft repo whose CRITICALs are mostly the scanner misreading infrastructure code.

https://skillscan.io/research/2026/two-scanners-one-corpus/2026-10-22T00:00:00Z

We ran NVIDIA's SkillSpector in static mode over every skill the Cisco scanner had already scanned. At the strict tier the two agree 80% of the time with kappa 0.26; at the loose tier kappa is 0.07. Where they disagree, and what a second verdict does and does not tell you.

https://skillscan.io/research/2026/618-skills-full-scan/2026-10-21T00:00:00Z

Results from scanning 618 distinct AI agent skills (650 scans) from ClawHub, skills.sh, and major skill repos with every analyzer enabled: 22% flagged, ClawHub far ahead, and most CRITICALs in a handful of security tools.

https://skillscan.io/research/2026/false-positive-register/2026-10-21T00:00:00Z

A maintained list of scanner findings on skillscan.io that we believe are wrong, with the reason, so CRITICAL counts can be read with the right discount.

https://skillscan.io/research/2026/clawhavoc-validation/2026-02-12T00:00:00Z

Koi AI found 341 malicious skills on ClawHub. Cross-referencing our scan results, Cisco's skill-scanner independently detected the same malware campaign — including the same C2 IP address and distribution patterns.

https://skillscan.io/research/2026/first-complete-scan/2026-02-08T00:00:00Z

After several iterations, API credit exhaustion, and pipeline fixes, we've completed our first clean run with full LLM and meta-analysis. Key insight: LLM analysis catches 83x more social engineering attacks than static analysis.

https://skillscan.io/research/2026/silent-api-failure-discovery/2026-02-05T00:00:00Z

We discovered that 540 of our scans silently failed to run LLM and meta-analysis due to exhausted API credits. The scanner reported success, but the analysis never happened. Here's how we found the bug and what the real results look like.

https://skillscan.io/research/2026/clawdefender-false-positives/2026-02-05T00:00:00Z

ClawDefender, a security skill for OpenClaw agents, was flagged with 37 findings and 22 CRITICAL severity -- making it the highest-finding-count skill in our dataset. Every single finding was a false positive.

https://skillscan.io/research/2026/why-we-built-this/2026-02-01T00:00:00Z

AI agent skills execute with full permissions and minimal oversight. An academic study found 26% contain vulnerabilities. We built a prototype automated scanning pipeline to explore the risk landscape.

https://skillscan.io/research/2026/first-scan-results/2026-02-01T00:00:00Z

An overview of our initial scan of the top 22 AI agent skills from skills.sh and clawhub.ai, revealing 3 unsafe skills and a dominant data exfiltration pattern.

https://skillscan.io/research/2026/static-vs-llm-scanning/2026-02-01T00:00:00Z

Comparing what static analysis catches versus what Claude Sonnet 4.5 finds -- and why the difference matters for AI agent security.

https://skillscan.io/research/2026/meta-analyzer-validation/2026-02-01T00:00:00Z

How the meta-analyzer reviews LLM findings for legitimacy, downgrades false alarms, and adds exploitability context.

https://skillscan.io/research/2026/scaling-to-62-skills/2026-02-01T00:00:00Z

Expanding from 22 to 62 (the top 50 from skills.sh plus our prior scans), skills revealed two new threat categories -- prompt injection and tool chaining abuse -- and surfaced browser-use as a high-risk skill class.