Security audit
Security checks for vulnerabilities and agentic risk
The skill’s research purpose is coherent, but it asks agents to fetch and run unpinned external code that can use a local Cue API key and consume credits.
Review this before installing if you are not comfortable with an agent downloading or updating third-party code under ~/.cue and running it with access to your Cue API key. Prefer a pinned, preinstalled, audited Cue runner, and confirm each credit-consuming run only for queries you intend to send to Cue.
The '何时用' section says only '深度核查：企业尽调、信披与监管', which is a broad domain description rather than a specific activation condition. It does not clarify what user requests should or should not invoke the skill, nor provide negative examples or scope constraints, increasing the risk of unintended invocation.
The skill instructs the agent to git pull or git clone and then rely on code from that external repository. This expands the skill from information retrieval into supply-chain and code-fetching behavior, allowing unreviewed remote content to influence later execution. In a skill file, that is dangerous because the fetched repository can change over time and may introduce malicious or unsafe logic without the user explicitly auditing it.
The skill directs the agent to execute a local Python script from the cloned repository, effectively granting subprocess and arbitrary code-execution capability. Because that script originates from externally fetched code, the agent is being asked to run code whose behavior is not constrained by the declared deep-verification purpose, creating a clear remote-code/supply-chain risk.
No suspicious patterns detected.
