byted-volcengine-rootcause-analyzer
Pass
Audited by Gen Agent Trust Hub on Jul 14, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection due to the ingestion of external, untrusted data to perform automated diagnostics.
- Ingestion points: The workflow processes 'alert text', 'log excerpts', and 'business symptom descriptions' from external monitoring systems and user conversations (SKILL.md).
- Boundary markers: Absent. The skill instructions do not specify the use of delimiters or escaping mechanisms to isolate untrusted data from agent instructions.
- Capability inventory: The skill has the authority to orchestrate multiple tools, including topology builders, cloud monitoring queries, and API assistants, to explore cloud resources (SKILL.md).
- Sanitization: Absent. There are no defined filters or validation logic to sanitize potentially malicious payloads embedded in the ingested logs or descriptions.
- [SAFE]: All referenced tools and dependencies (e.g., byted-volcengine-cloudmonitor) belong to the Volcengine product ecosystem, which is consistent with the author context (bytedance).
Audit Metadata