Lp3
Medium
- Category
- MCP Least Privilege
- Confidence
- 84% confidence
- Finding
- The skill advertises no declared permissions, yet its documented implementation characteristics imply access to environment variables, local files, file writes for caching, and outbound network requests. This creates a trust and review gap: agents or users may invoke the skill believing it is low-privilege when it can read local context, persist data, and fetch remote content, increasing the chance of unintended data exposure or policy bypass.
