Agent Skills Should Be Treated as Supply Chain Components
Overview
Tessl argues, in a talk at AI Native DevCon London reported on its blog, that agent skills should be treated as supply chain components because they can shape how AI agents behave.
Skills can be markdown instructions bundled with supporting files, so the author says reviewing the main SKILL.md file alone is not enough; risks can sit in supporting files, in later updates, and in workspace trust settings.
The author identifies the core danger as the combination of private context, untrusted content, and external communication. He cites research that scanned roughly 4,000 public skills and found issues including malware-like behavior. These are the author's and the cited research's claims as reported by Tessl, not independently verified findings.
Written by AI from the articles below · updated Oct 8, 8:02 PM ET
Check the sources:
Article timeline
Follow the coverage from different perspectives. Times are ET.
- Tessl BlogAgent Skills Should Be Treated as Supply Chain Components
AITessl's talk at AI Native DevCon London argues that agent skills, which can be markdown files with instructions and bundled material, act as supply chain components that can shape agent behavior. The author says reading SKILL.md once is insufficient because risks can sit in supporting files, updates, and workspace trust settings. He identifies the danger as the combination of private context, untrusted content, and external communication, and cites research scanning roughly 4,000 public skills for issues including malware-like behavior.
Heat trend
Not enough continuous observations to show a trend yet.