Scanned skills

JDArmy/BREAK

Checked the catalogued skill.

SkillTrust corpus scan 14 September 2026 v0.10.0 via ghsearch Verify on GitHub
F AI-triaged 12155
F Security

F; retained sample includes a medium finding · SD-008 · what this rule accepts as a false positive →

F Permissions

F set by findings not listed

B Transparency

B; retained sample includes a medium finding · SD-024 · what this rule accepts as a false positive →

showing 10 of 12155
transparencymedium SD-024
.mcp.json
securitymedium SD-008
CHANGELOG.md
securitymedium SD-008
CHANGELOG.md
securitymedium SD-008
CHANGELOG.md
securitymedium SD-008
CHANGELOG.md
securitymedium SD-008
CHANGELOG.md
securitymedium SD-008
CHANGELOG.md
securitymedium SD-008
CHANGELOG.md
securitymedium SD-008
CHANGELOG.md
securitymedium SD-008
CHANGELOG.md
4 suppressed by AI triage
securitymedium SD-012 benign_example
package.json

The prepare script simply installs Husky for Git hooks, a standard and benign usage of a lifecycle hook.

transparencymedium SD-007 benign_example
src/BREAK/attack-tools/AT0017.json

The file contains only a static URL reference, not an executed network call or permission grant.

transparencymedium SD-007 benign_example
src/BREAK/avoidances/A0136.json

Static reference URL in JSON, not an actual outbound call.

transparencymedium SD-007 benign_example
src/BREAK/avoidances/A0161.json

The URL appears only as a reference in a metadata file, not as an executed outbound network call.