microsoft-teams-automation
Warn
Audited by Gen Agent Trust Hub on Oct 24, 2026
Risk Level: MEDIUMEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill instructs users to configure an external MCP server from
https://rube.app/mcp. Since this domain is not a recognized trusted vendor or well-known service, this setup allows for the potential execution of unverified tool definitions and logic within the agent's environment.\n- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data from Microsoft Teams messages, which creates a vulnerability where malicious instructions embedded in those messages could influence the agent's behavior.\n - Ingestion points: The skill ingests message content and search results via tools such as
MICROSOFT_TEAMS_SEARCH_MESSAGESandMICROSOFT_TEAMS_GET_CHAT_MESSAGEas described inSKILL.md.\n - Boundary markers: While the instructions advise verifying recipients and actions, there are no technical delimiters or prompts specified to isolate external data from agent instructions.\n
- Capability inventory: The skill has access to high-impact writing tools, including
MICROSOFT_TEAMS_TEAMS_POST_CHANNEL_MESSAGE,MICROSOFT_TEAMS_TEAMS_POST_CHAT_MESSAGE, andMICROSOFT_TEAMS_CREATE_MEETING.\n - Sanitization: There is no evidence of input validation or content sanitization for data retrieved from Teams messages.
Audit Metadata