Security audit
Security checks for vulnerabilities and agentic risk
This feed-digest skill has no embedded malware, but it asks for broad read/write/exec authority and describes generic automation beyond feed summaries without tight scoping.
Review this skill before installing. It may be usable for feed summaries only in a tightly sandboxed agent, but do not grant broad shell execution or write access unless you are comfortable with generic automation authority and can enforce command, file, and network limits outside the skill.
The invocation guidance uses broad trigger language such as recommending use whenever AI models, agent orchestration, or LLM applications are involved, without defining clear activation boundaries. In agent systems, overly broad routing criteria can cause the skill to be invoked in unrelated contexts, exposing read/write/exec tools more often than necessary and increasing the attack surface.
Advertising system command execution in a feed-digest skill is dangerous because command execution is a powerful primitive unrelated to the core use case. In an agent environment, this can normalize or trigger shell access where only network fetch and content processing should be needed, enabling abuse if later paired with user-controlled input or ambiguous instructions.
User-facing fields such as displayName, summary, and description combine Chinese and English, but the file does not state whether output language follows user preference or whether bilingual behavior is optional. This can create an implicit language-policy issue because the skill appears to impose mixed-language UX without user opt-in.
The documentation creates a misleading trust boundary by calling the skill a straightforward feed helper while elsewhere framing it as a broader automation/execute skill. This inconsistency can cause operators to underestimate the privilege and risk profile of the skill, leading to overbroad deployment or approval.
The skill is presented as a feed-digest helper, but its documented capabilities expand into generic file handling, API integration, command execution, and information retrieval. This scope drift weakens least-privilege expectations and can cause an agent or user to authorize broader actions than are necessary for feed summarization, increasing the chance of misuse or unsafe tool invocation.
