Context-Inappropriate Capability
Medium
- Confidence
- 88% confidence
- Finding
- The skill instructs the agent to retrieve a bearer token from environment variables and OS credential stores, which expands the agent's capability from answering questions into accessing local secrets. Although the token is intended for the daily.dev API, granting the skill secret-access behavior is risky because any prompt injection, logging, error handling, or future modification could expose the token or normalize unnecessary credential access.
