score-analyzer
Pass
Audited by Gen Agent Trust Hub on Oct 22, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill instructions and examples focus exclusively on text processing and statistical reporting. It does not perform network operations, access sensitive files, or execute system commands.
- [INDIRECT_PROMPT_INJECTION]: The skill possesses a surface for indirect prompt injection as it ingests external scoring data in
SKILL.md("接收多轮评估结果"). However, the risk is negligible due to the following:- Ingestion points: User-provided structured text containing evaluation scores.
- Boundary markers: Absent.
- Capability inventory: The skill has no capabilities to execute shell commands, write files, or perform network requests.
- Sanitization: Not explicitly defined, but unnecessary given the lack of dangerous tools.
Audit Metadata