Context-Inappropriate Capability
Medium
- Confidence
- 92% confidence
- Finding
- The skill instructs the agent to execute a shell script (`./scripts/init-novel.sh`) when certain conditions are met, even though the skill is primarily for content generation. Any instruction that can trigger command execution expands the attack surface: a user could steer the agent into running local scripts in a workspace the analyst has not validated, and the script itself could perform unintended filesystem or environment actions.
