Security audit
Security checks for vulnerabilities and agentic risk
This employee-monitoring skill is mostly coherent, but it silently creates/reuses service identities, stores tokens locally, and sends workplace media and report history to a cloud service with limited user-facing control.
Install only in an authorized workplace-monitoring setting after legal/privacy review. Expect employee images or videos, media URLs, identity-linked report history, and generated reports to be handled by the configured Life Emergence cloud APIs, and require clear employee notice, service-account controls, token storage review, and retention/deletion procedures before use.
Automatic authentication, account creation/lookup, local workspace access, and default open-id generation are materially beyond the declared purpose of analyzing media for phone use. Hidden identity provisioning tied to remote APIs can cause unauthorized account creation, silent cross-session linkage, and privacy-sensitive report access without informed user consent.
Automatic authentication, account creation/lookup, local workspace access, and default open-id generation are materially beyond the declared purpose of analyzing media for phone use. Hidden identity provisioning tied to remote APIs can cause unauthorized account creation, silent cross-session linkage, and privacy-sensitive report access without informed user consent.
Automatic authentication, account creation/lookup, local workspace access, and default open-id generation are materially beyond the declared purpose of analyzing media for phone use. Hidden identity provisioning tied to remote APIs can cause unauthorized account creation, silent cross-session linkage, and privacy-sensitive report access without informed user consent.
Automatic authentication, account creation/lookup, local workspace access, and default open-id generation are materially beyond the declared purpose of analyzing media for phone use. Hidden identity provisioning tied to remote APIs can cause unauthorized account creation, silent cross-session linkage, and privacy-sensitive report access without informed user consent.
Automatic authentication, account creation/lookup, local workspace access, and default open-id generation are materially beyond the declared purpose of analyzing media for phone use. Hidden identity provisioning tied to remote APIs can cause unauthorized account creation, silent cross-session linkage, and privacy-sensitive report access without informed user consent.
Automatic authentication, account creation/lookup, local workspace access, and default open-id generation are materially beyond the declared purpose of analyzing media for phone use. Hidden identity provisioning tied to remote APIs can cause unauthorized account creation, silent cross-session linkage, and privacy-sensitive report access without informed user consent.
Automatic authentication, account creation/lookup, local workspace access, and default open-id generation are materially beyond the declared purpose of analyzing media for phone use. Hidden identity provisioning tied to remote APIs can cause unauthorized account creation, silent cross-session linkage, and privacy-sensitive report access without informed user consent.
Automatic authentication, account creation/lookup, local workspace access, and default open-id generation are materially beyond the declared purpose of analyzing media for phone use. Hidden identity provisioning tied to remote APIs can cause unauthorized account creation, silent cross-session linkage, and privacy-sensitive report access without informed user consent.
Automatic authentication, account creation/lookup, local workspace access, and default open-id generation are materially beyond the declared purpose of analyzing media for phone use. Hidden identity provisioning tied to remote APIs can cause unauthorized account creation, silent cross-session linkage, and privacy-sensitive report access without informed user consent.
Automatic authentication, account creation/lookup, local workspace access, and default open-id generation are materially beyond the declared purpose of analyzing media for phone use. Hidden identity provisioning tied to remote APIs can cause unauthorized account creation, silent cross-session linkage, and privacy-sensitive report access without informed user consent.
Automatic authentication, account creation/lookup, local workspace access, and default open-id generation are materially beyond the declared purpose of analyzing media for phone use. Hidden identity provisioning tied to remote APIs can cause unauthorized account creation, silent cross-session linkage, and privacy-sensitive report access without informed user consent.
Automatic authentication, account creation/lookup, local workspace access, and default open-id generation are materially beyond the declared purpose of analyzing media for phone use. Hidden identity provisioning tied to remote APIs can cause unauthorized account creation, silent cross-session linkage, and privacy-sensitive report access without informed user consent.
YARA rule matched a hack tool or exploit indicator (offensive tools, reconnaissance, privilege escalation, or exploit frameworks).
---
name: "phone_usage_monitoring_analysis"
description: "Based on computer vision, automatically detects employees playing with phones during work hours, supports real-time video stream and image detection, counts the duration and frequency of phone usage, helps enterprises standardize office order, and improves work efficiency. | 职场玩手机智能监测技能，基于计算机视觉自动检测工作时间员工玩手机行为，支持视频流和图片实时检测，统计玩手机时长与频次，帮助企业规范办公秩序，提升工作效率"
version: "1.0.20"
license: "MIT-0"
---
# 📱 Workplace Phone Usage Smart Monit
The skill centers on employee surveillance using images/video and cloud report access, yet it does not provide prominent upfront notice about privacy-sensitive data handling, retention, remote transmission, or identity linkage. In this context, lack of disclosure is especially dangerous because the processed data may include employees, workplace activity, and account-scoped reports.
The code provides arbitrary remote resource modification capability through generic add/edit/delete and raw POST/PUT/DELETE wrappers without visible constraints on destination or operation. Because the declared skill purpose is passive monitoring/analysis, write-capable network primitives are unjustified and could be abused to alter external systems, especially if other parts of the skill can influence the URL or payload.
The User model stores sensitive identity and authentication-related fields including username, realname, email, token, and open_token, yet the declared skill purpose is workplace phone-use detection. Unjustified retention of credentials or tokens materially raises privacy and account-compromise risk, especially if the local SQLite database is accessed by other components or users.
The utility layer automatically provisions identities via a remote '/sys/phoneLogin' call, retrieves tokens, and persists user/token material locally, even though the skill is presented as phone-usage monitoring. This is dangerous because it silently expands scope from vision analysis into account creation, authentication, and credential storage, creating privacy, consent, and token-compromise risks unrelated to the stated function.
The skill declares no explicit tool scope or permission boundaries despite instructing use of shell execution, local file handling, environment-backed identity initialization, and network/API access. In an agent environment, missing tool constraints can let the skill invoke broader capabilities than users expect, increasing the blast radius if the skill is triggered unintentionally or combined with unsafe inputs.
Broad workplace-monitoring trigger keywords can cause the skill to activate on loosely related conversations and automatically process uploaded media or initiate surveillance-oriented actions. In a privacy-sensitive employee-monitoring context, unintended activation increases the chance of unnecessary analysis, network transmission, or persistence of sensitive data.
The history-report auto-trigger phrases are broad enough that casual requests about reports could silently invoke cloud queries. That can disclose metadata about prior monitoring sessions or cause account-linked remote access without the user clearly consenting to a backend lookup.
The skill states that identity parameters are handled internally, may reuse a local default user, and may auto-create one if absent. For a workplace-surveillance skill, silent identity initialization enables covert account linkage and persistence of sensitive report access under an opaque identity, making misuse and unauthorized data access more dangerous.
The documentation states that a cloud visual AI service analyzes workplace images/video of employees, but it does not explicitly warn users that sensitive employee imagery is transmitted to a remote service for processing. In an employee surveillance context, this materially increases privacy, consent, and compliance risk because operators may deploy the system without adequate notice, lawful basis, data handling disclosures, or cross-border transfer assessment.
User-facing strings throughout the script, including errors, prompts, help text, and reports, are fixed in Chinese. The file does not provide a language/locale option or explain that the skill is intentionally limited to Chinese-speaking users.
The function exposes a capability to retrieve prior analysis records for an OpenID-backed internal identity, which goes beyond the declared purpose of analyzing a supplied image or video. In an employee-monitoring context, that can reveal historical behavioral surveillance data and creates an access-scope expansion risk if authorization is weak or implicit.
The code silently resolves and uses an internal OpenID identity context even though the skill presents itself as a media-analysis tool. Hidden identity binding can cause data to be fetched or attributed under an internal user context without clear disclosure, increasing the risk of unintended access to employee-monitoring records.
Detected: suspicious.install_untrusted_source