ceae-skill
Pass
Audited by Gen Agent Trust Hub on Aug 27, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill provides a structured framework for hardware engineers to perform root cause analysis. The instructions are focused on engineering logic (5Why, TRIZ, First Principles) and do not contain any prompt injection or bypass attempts.
- [COMMAND_EXECUTION]: The skill utilizes a local script,
scripts/ceae_tree_export.py, to generate PNG因果图 (causal diagrams). This script uses well-known, standard Python libraries (matplotlib,networkx) and includes text sanitization logic (_strip_emoji) to handle user-provided input safely during the visualization process. It does not perform any unauthorized file system or network operations. - [EXTERNAL_DOWNLOADS]: The skill references
open.zhihuiya.comfor MCP (Model Context Protocol) configuration. This is the official platform domain for the skill's author, PatSnap (Zhihuiya is the Chinese brand name for PatSnap). References to official vendor resources for configuration purposes are considered safe and standard practice. - [INDIRECT_PROMPT_INJECTION]: While the skill ingests fuzzy engineering targets from users, the risk is low due to the structured processing steps (parameterization, formula breakdown, and logic-based 5Why analysis) which prevent arbitrary user input from influencing the agent's core safety or execution environment.
Audit Metadata