cloud-api-operations

Pass

Audited by Gen Agent Trust Hub on Sep 9, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill fetches API documentation, index files, and SDK schema definitions from official Tencent Cloud domains (cloud.tencent.com, docs.cloudbase.net) and their verified source code repositories on GitHub (github.com/TencentCloud), Gitee, and CNB. These resources are used for API action and parameter validation.
  • [COMMAND_EXECUTION]: Utilizes platform-provided tools auth and callCloudApi to manage environment credentials and perform cloud resource operations. It specifically includes a flow to retrieve temporary credentials for use in user scripts.
  • [INDIRECT_PROMPT_INJECTION]: The agent is instructed to ingest and parse external content, including the API reference index and SDK source files. This creates a vulnerability surface where malformed or malicious content in these external sources could attempt to influence the agent's behavior.
  • Ingestion points: SKILL.md (Workflow Step 1) and references/calling-methods.md (§2) describe fetching data from docs.cloudbase.net and TencentCloud's GitHub/CNB/Gitee repositories.
  • Boundary markers: None explicitly defined in the provided instructions for these ingestion points.
  • Capability inventory: The agent has access to callCloudApi for control-plane operations and auth for credential retrieval.
  • Sanitization: No specific sanitization or filtering logic is described for the ingested documentation or source code.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 9, 2026, 08:42 AM
Security Audit — agent-trust-hub — cloud-api-operations