Security audit
Security checks for vulnerabilities and agentic risk
This is a coherent Chinese-language sales assistant skill that handles sensitive sales records, but its data use and human-review boundaries are disclosed and purpose-aligned.
Install only in an environment where Feishu access, Hermes Agent skills, CRM or archive locations, audit logs, retention periods, and deletion procedures are explicitly configured. Users should avoid uploading real customer, contract, employee, financial, or technical-sensitive data until masking, authorization checks, and human approval routing are confirmed.
The manifest description, headings, workflow instructions, and default output format are all written to operate in Chinese, and the template at L74-L90 defines Chinese-only output labels. There is no indication that users may choose another language or that the skill is intentionally limited to a Chinese-language or region-specific deployment.
The file's natural-language instructions and example invocation are entirely in Chinese, with no indication that the user can choose another language or that the skill is intentionally restricted to a Chinese-speaking context. Under the stated policy, forcing a specific language without user opt-in can constitute a locale/language policy violation.
The file is entirely written as a mandatory Chinese-language knowledge base description for the skill, with no indication that users may choose another language or that the skill is intentionally limited to a Chinese-only regional/compliance context. Under the policy, forcing a specific language without opt-in can be a natural-language policy violation.
The manifest presents all user-facing skill names and purpose descriptions exclusively in Chinese across the defined skills. Because there is no stated opt-in, alternative locale, or justification that this is a China-specific deployment, this creates a natural-language locale constraint that may violate language-choice policy.
The entire system prompt is written as mandatory Chinese-language operating instructions, and there is no indication that users may choose another language or locale. This can violate language/locale policy when a skill enforces a specific language without user opt-in or documented regional justification.
The file content is entirely in Chinese and does not indicate that language selection is optional or limited to a justified region-specific context. This can violate a language/locale policy when users are not given an explicit opt-in or alternative language option.
The entire report template is written in Chinese, which implies the skill's generated output will be in a specific language by default. The file does not indicate that language selection is optional or justified by a region-specific requirement, which conflicts with the policy against forcing a locale without user opt-in.
The template title is written entirely in Chinese, indicating the skill content is constrained to a specific language. The file does not offer a language choice or explain that the template is intended only for a Chinese-language or region-specific workflow, which creates a natural-language policy concern under the locale/language rule.
The file presents the skill entirely in Chinese, including the invocation example and expected outputs, with no indication that users may choose another language or locale. This can violate language-choice policy when a skill appears to require a specific language without explicit opt-in or documented locale constraints.
This workflow explicitly ingests user messages and attachments, performs knowledge-base retrieval and analysis, and then logs and archives the results, but it does not provide a clear user-facing disclosure about how potentially sensitive customer and business data will be processed and retained. In a sales-assistant context, this creates privacy and compliance risk because users may submit confidential customer information without informed consent or awareness of storage and downstream handling.
The customer profiling workflow analyzes uploaded customer materials, extracts structured fields, retrieves historical materials, and archives outputs, yet it does not warn that sensitive business information may be correlated across sources and persisted. Because customer profiling often includes organizational charts, budgets, decision chains, and contact details, undisclosed aggregation increases confidentiality and compliance exposure.
This workflow processes visit records, chat logs, and audio transcripts into meeting minutes and archives them, but it does not disclose that these sources may contain sensitive personal data, confidential business discussions, or third-party information. That omission is risky because transcription and summarization can normalize broad collection of sensitive content without ensuring the submitter has authority or consent to process and retain it.
The customer archiving workflow is designed to persist customer records, project stages, key contacts, and next steps, but it does not explicitly inform users that personally identifiable and commercially sensitive information will be retained in archives. In this skill context, persistent CRM-like storage increases the risk of unauthorized retention, over-collection, and regulatory noncompliance if users are not warned and safeguards are not stated.
The demo script explicitly states that results will be archived, but it does not mention any warning, consent step, retention period, or data-handling constraints. In a sales-assistant context, archived outputs may contain customer visit notes, profiles, needs, and risk indicators, so silent retention can expose sensitive business or personal information beyond the immediate task.
The manifest presents the skill identity, deployment channel, and mission entirely in Chinese, which indicates a fixed language/locale experience. There is no explicit opt-in, language selection, or documentation that this skill is intentionally limited to a Chinese-speaking region or audience.
This markdown file contains only Chinese-language operational instructions, and there is no indication that users may select another language or that the skill is intentionally limited to a Chinese-speaking audience. Under the policy, forcing a specific language without user opt-in is a natural-language policy concern.
The template heading is written in Chinese ("销售助手归档记录") with no indication that language selection is optional or tied to a region-specific use case. This can violate language/locale policy when a skill implicitly enforces a specific language without user opt-in.
This markdown template presents the skill content only in Chinese, which can amount to a language-policy issue when no user opt-in or alternative language is offered. The file does not indicate that the template is intended solely for a Chinese-language or region-specific workflow.
The workflow states it can generate follow-up actions and Feishu reminders, but it does not clearly warn users that downstream communications or records may be created based on analyzed content. While lower risk than full archival of raw sensitive content, it can still cause privacy, audit, and operational issues if users do not understand that the system may create persistent reminders or workflow artifacts tied to customer interactions.
No suspicious patterns detected.
