cloud-api-operations
Pass
Audited by Gen Agent Trust Hub on Oct 9, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSPRIVILEGE_ESCALATIONOBFUSCATIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill instructs the agent to fetch SDK source code from official Tencent Cloud repositories on GitHub, Gitee, and CNB. This is used to accurately discover parameter schemas for API calls and is limited to trusted vendor resources.
- [PRIVILEGE_ESCALATION]: The skill facilitates the generation of CAM authorization links for users to update service role permissions. It includes explicit warnings about the impact of these changes and recommends manual intervention for high-risk operations like financial transactions.
- [OBFUSCATION]: Base64 encoding is employed for formatting the
principalparameter within vendor-specific authorization URLs, adhering to the official Tencent Cloud CAM API specifications. - [INDIRECT_PROMPT_INJECTION]: The skill possesses a potential attack surface by processing external data from cloud APIs and SDK files. It mitigates this risk by requiring cross-verification with official documentation and providing clear instructions for error classification and recovery.
Audit Metadata