mouse-swarm-reveal
Fail
Audited by Snyk on Nov 13, 2026
Risk Level: CRITICAL
Full Analysis
CRITICAL E004: Prompt injection detected in skill instructions.
- Potential prompt injection detected (medium risk: 0.30). The skill's visual reveal and status-view code is legitimate and self-described, but the
npx --yesline runs unpinned GitHub package code without vetting, which is risky toward the user even though it is documented and not clearly malicious.
MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).
- Third-party content exposure detected (low risk: 0.10). The required Ruflo observation path polls agent registrations and statuses from the local Ruflo CLI (SKILL.md:56), which is the user's own system state; an outsider has no direct submission path into it, but agent labels and roles are free text that could be shaped indirectly, so this is rated low under recall-first.
MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).
- Potentially malicious external URL detected (medium risk: 0.30). The skill's setup instructs
npm install github:ruvnet/mouse-swarmandnpx --package=github:ruvnet/mouse-swarm, which fetch and execute unpinned code from an individual-owned GitHub repository, so it is an unofficial dependency with no concrete warning sign and no strong first-party provenance, giving a medium tier.
Issues (3)
E004
CRITICALPrompt injection detected in skill instructions.
W011
MEDIUMThird-party content exposure detected (indirect prompt injection risk).
W012
MEDIUMUnverifiable external dependency detected (runtime URL that controls agent).
Audit Metadata