Security audit
Security checks for vulnerabilities and agentic risk
This skill is a disclosed interview-assistant workflow that handles sensitive hiring data, but I found no hidden execution, exfiltration, persistence, or destructive behavior.
Install only if you are authorized to process candidate resumes, contact details, meeting recordings, transcripts, and interview notes. Use the generated scoring as decision support, not as the sole basis for hiring, and apply your organization's privacy, consent, retention, and bias-review requirements.
The skill claims end-to-end interview support using Tencent Meeting directory and transcript data, but the described implementation does not substantiate those integrations or capabilities. This mismatch can mislead users into believing the skill has validated access paths, controls, and coverage it may not actually have, increasing the risk of unsafe handling of candidate data and incorrect operational decisions.
The skill clearly handles highly sensitive personal data: candidate resumes, contact information, meeting transcripts, interview notes, and generates hiring evaluations from them. Omitting an explicit warning and consent/privacy notice increases the risk that users process protected HR data without understanding retention, access, or compliance implications, which can lead to privacy violations and improper employment decision-making.
Without declared permissions the skill's intent is opaque and cannot be validated.
Broad trigger phrases can cause the skill to activate in ambiguous situations and process sensitive hiring-related content without clear boundaries. In a recruiting context, overbroad invocation increases the chance of accessing or summarizing candidate records and meeting notes when the user did not intend this specific skill path.
The skill proposes accessing candidate directory information and meeting transcripts without an explicit user warning or consent checkpoint. Because these sources contain sensitive personal and interview data, silent or implicit access can lead to privacy violations, unauthorized data use, and noncompliance with hiring-data handling requirements.
The entire document is generated in Chinese, and L16 states it was produced by the skill, but there is no indication that the user opted into Chinese output or that the skill is limited to a Chinese-language or region-specific workflow. This can violate language/locale policy when a skill imposes a specific language by default.
The module docstring and all user-facing descriptions specify the skill entirely in Chinese and frame the output as a Chinese-language interview evaluation form, without offering any language/locale choice. This is a natural-language locale policy concern because the skill appears to impose a specific language on users rather than making it configurable or clearly documenting a justified region-specific constraint.
The manifest defines activation entirely through Chinese phrases and does not indicate that other languages are supported or that Chinese is an intentional locale-limited requirement. This can violate language/locale policy when a skill implicitly requires one language without user opt-in or justification.
This markdown file contains user-facing natural-language content entirely in Chinese and does not indicate that the language is optional or user-selectable. Under the stated policy, forcing a specific language without user opt-in can be a natural-language policy violation.
The file’s title, headings, and example evaluation content are entirely in Chinese, which indicates the skill template is designed to operate in a fixed language. The policy requires flagging language or locale constraints when they are imposed without user opt-in or an explicitly justified regional scope.
