Security audit
Security checks for vulnerabilities and agentic risk
This skill is a cloud-backed workplace surveillance analyzer that mostly matches its stated purpose, but it silently creates or reuses identity records, stores tokens locally, and sends sensitive media and identity data to external services with insufficient disclosure.
Review before installing. Only use this skill if you are comfortable sending workplace images/videos or media URLs, plus an internal user identity, to the lifeemergence.com/open.lifeemergence.com services. Ask the publisher for clear retention, authorization, account-creation, token-storage, and report-history access terms, and prefer a version that requires explicit confirmation before media upload or history queries.
The documented scope is narrow operational monitoring, but the code indicators include external HTTP communication, authentication/token handling, automatic account creation/login, workspace discovery, and local identity persistence. These capabilities materially increase attack surface and privacy risk because they enable remote service interaction and durable user correlation under a misleading label.
The documented scope is narrow operational monitoring, but the code indicators include external HTTP communication, authentication/token handling, automatic account creation/login, workspace discovery, and local identity persistence. These capabilities materially increase attack surface and privacy risk because they enable remote service interaction and durable user correlation under a misleading label.
The documented scope is narrow operational monitoring, but the code indicators include external HTTP communication, authentication/token handling, automatic account creation/login, workspace discovery, and local identity persistence. These capabilities materially increase attack surface and privacy risk because they enable remote service interaction and durable user correlation under a misleading label.
The documented scope is narrow operational monitoring, but the code indicators include external HTTP communication, authentication/token handling, automatic account creation/login, workspace discovery, and local identity persistence. These capabilities materially increase attack surface and privacy risk because they enable remote service interaction and durable user correlation under a misleading label.
The documented scope is narrow operational monitoring, but the code indicators include external HTTP communication, authentication/token handling, automatic account creation/login, workspace discovery, and local identity persistence. These capabilities materially increase attack surface and privacy risk because they enable remote service interaction and durable user correlation under a misleading label.
The documented scope is narrow operational monitoring, but the code indicators include external HTTP communication, authentication/token handling, automatic account creation/login, workspace discovery, and local identity persistence. These capabilities materially increase attack surface and privacy risk because they enable remote service interaction and durable user correlation under a misleading label.
The documented scope is narrow operational monitoring, but the code indicators include external HTTP communication, authentication/token handling, automatic account creation/login, workspace discovery, and local identity persistence. These capabilities materially increase attack surface and privacy risk because they enable remote service interaction and durable user correlation under a misleading label.
The documented scope is narrow operational monitoring, but the code indicators include external HTTP communication, authentication/token handling, automatic account creation/login, workspace discovery, and local identity persistence. These capabilities materially increase attack surface and privacy risk because they enable remote service interaction and durable user correlation under a misleading label.
The documented scope is narrow operational monitoring, but the code indicators include external HTTP communication, authentication/token handling, automatic account creation/login, workspace discovery, and local identity persistence. These capabilities materially increase attack surface and privacy risk because they enable remote service interaction and durable user correlation under a misleading label.
The documented scope is narrow operational monitoring, but the code indicators include external HTTP communication, authentication/token handling, automatic account creation/login, workspace discovery, and local identity persistence. These capabilities materially increase attack surface and privacy risk because they enable remote service interaction and durable user correlation under a misleading label.
The documented scope is narrow operational monitoring, but the code indicators include external HTTP communication, authentication/token handling, automatic account creation/login, workspace discovery, and local identity persistence. These capabilities materially increase attack surface and privacy risk because they enable remote service interaction and durable user correlation under a misleading label.
The documented scope is narrow operational monitoring, but the code indicators include external HTTP communication, authentication/token handling, automatic account creation/login, workspace discovery, and local identity persistence. These capabilities materially increase attack surface and privacy risk because they enable remote service interaction and durable user correlation under a misleading label.
YARA rule matched a hack tool or exploit indicator (offensive tools, reconnaissance, privilege escalation, or exploit frameworks).
---
name: "staff-absence-detection-analysis"
description: "Real-time monitoring of personnel on-duty status in specific areas based on computer vision and human pose estimation, automatically detects abnormal statuses such as leaving posts and absent from work, supports custom threshold settings, and triggers early warning immediately when abnormality is detected. | 人员离岗实时监测技能，基于计算机视觉与人体姿态估计算法，实时监测特定区域内人员的在岗状态，自动判断离岗、缺岗等异常状态，支持自定义判定阈值，异常发生立即触发预警，适用于工厂车间、监控室
The fallback behavior creates persistent local user identities and stores them for later API use without a clear business need tied to absence detection. This expands the skill into identity creation and persistence, which can enable unauthorized account sprawl, attribution confusion, and hidden external service usage.
The utility layer performs token bootstrap, account lookup, and remote account provisioning/login flows that are unrelated to the stated staff-absence detection purpose. This creates an unexpected identity-management and outbound-authentication capability, increasing the chance of covert data transmission, account abuse, and use of the host environment's stored identity material.
The skill documents behaviors that require shell, filesystem, network, and environment access, but it does not declare any explicit tool scope or permissions boundaries. This creates an overbroad and opaque execution surface where the agent may invoke powerful capabilities without user-visible constraints, increasing the risk of unintended file access, command execution, or network egress.
The manifest framing emphasizes real-time monitoring, while the documentation also adds historical cloud report retrieval and report-link browsing. Undeclared secondary capabilities can surprise users and reviewers, especially when they involve remote data access and potentially sensitive past reports.
The trigger conditions are broad enough to auto-activate on generic references to monitoring or report history, which can cause unintended execution. In a skill that can save files locally and contact cloud APIs, accidental activation increases the chance of unnecessary data processing or disclosure.
The skill includes cloud historical report querying and persistent user identity handling that exceed the stated real-time analysis purpose. This scope expansion matters because it introduces privacy-sensitive longitudinal data access and identity correlation without clear need or narrowly defined authorization boundaries.
The skill states that uploaded attachments are automatically saved as local files without a clear user warning. Silent local persistence of user-provided media can create privacy, retention, and forensic exposure risks, especially for surveillance footage or workplace images.
The documentation does not clearly warn that historical report queries use a cloud API and that network media URLs may be fetched remotely. This matters because users may unintentionally send sensitive operational media or query metadata to third-party services without informed consent.
The documentation states that personnel images/videos and an open_id are sent to a cloud API, but it does not clearly disclose the privacy and security implications of transmitting potentially sensitive workplace surveillance data to a remote service. In this skill context, the data involves employee monitoring and status analysis, which makes omission of data-handling, retention, and consent guidance more dangerous than a generic API integration issue.
The method returns a fixed Chinese string for analysis output, with no indication that users can choose their preferred language or locale. This creates a natural-language policy concern because the skill appears to enforce one language unconditionally.
The script's natural-language strings, CLI descriptions, errors, and status messages are consistently hardcoded in Chinese. This forces a specific language for all users without offering a language selection or documenting a justified locale restriction, which matches the language/locale policy violation criteria.
The script exposes behavior beyond the advertised absence-detection function by allowing history listing tied to an internal user identity (show_analyze_list) and by resolving an OpenID in a non-transparent way. In a monitoring/surveillance context, hidden identity-linked retrieval increases privacy risk and can enable access to prior analysis data without clear user awareness or authorization boundaries.
Detected: suspicious.install_untrusted_source