Security Audit
outlook-calendar-automation
github.com/ComposioHQ/awesome-claude-skillsTrust Assessment
outlook-calendar-automation received a trust score of 95/100, placing it in the Trusted category. This skill has passed all critical security checks and demonstrates strong security practices.
SkillShield's automated analysis identified 1 finding: 0 critical, 0 high, 1 medium, and 0 low severity. Key findings include Potential for unauthorized access to other users' calendars.
The analysis covered 4 layers: Manifest Analysis, Static Code Analysis, Dependency Graph, LLM Behavioral Safety. All layers scored 70 or above, reflecting consistent security practices.
Last analyzed on February 17, 2026 (commit 99e2a295). SkillShield performs automated 4-layer security analysis on AI skills and MCP servers.
Layer Breakdown
Behavioral Risk Signals
Security Findings1
| Severity | Finding | Layer | Location | |
|---|---|---|---|---|
| MEDIUM | Potential for unauthorized access to other users' calendars The skill description indicates that tools like `OUTLOOK_LIST_CALENDARS` and `OUTLOOK_CALENDAR_CREATE_EVENT` can operate on 'other users' calendars' by specifying their email or UUID in the `user_id` parameter. While this functionality might be intended for legitimate shared calendar management, it introduces a risk where an LLM could be prompted to access or modify calendars of individuals other than the authenticated user, potentially without the explicit consent of the current LLM user for that specific action. This capability relies on broad underlying OAuth scopes (e.g., `Calendars.ReadWrite.Shared` or `Calendars.ReadWrite.All`) which, if granted, could be misused by a malicious prompt. 1. **Restrict OAuth Scopes**: Ensure the underlying Rube/Composio connection only requests the minimum necessary OAuth scopes. If shared calendar management is not a primary use case, avoid broad scopes like `Calendars.ReadWrite.Shared` or `Calendars.ReadWrite.All`. 2. **User Confirmation**: Implement explicit user confirmation prompts within the agent's workflow whenever an action is requested for a `user_id` other than 'me'. 3. **Policy Enforcement**: Enforce policies at the Rube/Composio platform level to prevent unauthorized access to other users' calendars, even if the token technically allows it. | LLM | SKILL.md:60 |
Scan History
Embed Code
[](https://skillshield.io/report/55d4b6d90b238e93)
Powered by SkillShield