Context-Inappropriate Capability
Medium
- Confidence
- 93% confidence
- Finding
- The skill tells the agent/user to install a third-party skill via `npx molthub@latest install skill-name` while performing a review task, which exceeds the minimum access needed to inspect a skill and introduces unnecessary code-fetching and execution risk. In a security-review context, normalizing installation of untrusted packages or skills is especially risky because the target content is itself potentially adversarial.
