alicloud-ai-cloud-call-center-test
Pass
Audited by Gen Agent Trust Hub on Mar 11, 2026
Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [PROMPT_INJECTION]: The skill exhibits a surface for indirect prompt injection by processing external data to drive its actions. Ingestion points: The agent reads the 'SKILL.md' file from a target skill directory to identify API calls. Boundary markers: Absent; there are no delimiters or instructions to ignore embedded commands within the ingested file. Capability inventory: The skill can execute local Python scripts and make network API calls. Sanitization: Absent; the skill does not validate the API names or parameters extracted from the external file before use.
- [COMMAND_EXECUTION]: The skill executes a local Python script 'tests/common/compile_skill_scripts.py' to validate the compilation of target skill scripts.
Audit Metadata