The Agent Skills Directory

[SAFE]: The skill's primary analysis is performed by a local Python script (scripts/deai.py) that relies exclusively on the standard library. It contains no networking, dynamic execution, or unauthorized file-writing capabilities.
[SAFE]: The instructions in SKILL.md and references/平台标注.md explicitly enforce security and ethical boundaries, including the 'Five Red Lines' which prohibit bypassing AI markers or aiding in academic dishonesty.
[INDIRECT_PROMPT_INJECTION]: As the skill processes untrusted user-provided text for analysis and rewriting, it has an inherent surface for indirect prompt injection. However, the risk is mitigated as the skill does not possess high-privilege capabilities such as network access or shell command execution, limiting the potential impact of such an attack.
[COMMAND_EXECUTION]: Documentation in references/draft/draft-changes.md mentions interacting with a separate developer tool (dby-api) via a local Node.js script. This is presented as an optional integration for project management and does not exhibit signs of malicious intent or command injection vulnerabilities.

dby-deai