v1.0.2

fugui-monitor

ReviewClawScan verdict for this skill. Analyzed May 1, 2026, 7:00 AM.

Analysis

The skill mostly matches its monitoring purpose, but it uses a logged-in browser session and sends alerts to a hardcoded Feishu recipient, so it should be reviewed before installation.

GuidanceBefore installing, verify who owns the hardcoded Feishu recipient ID and change it to your own configurable target. Use a dedicated Xiaohongshu browser profile if possible, and only enable the cron schedule if you want continuous monitoring.

Findings (5)

Artifact-based informational review of SKILL.md, metadata, install specs, static scan signals, and capability signals. ClawScan does not execute the skill or run runtime probes.

Abnormal behavior control

Checks for instructions or behavior that redirect the agent, misuse tools, execute unexpected code, cascade across systems, exploit user trust, or continue outside the intended task.

Rogue Agents

SeverityLowConfidenceHighStatusNote

SKILL.md

08:00-18:00 每5分钟，18:00-24:00 每10分钟

The skill is designed for recurring background checks via cron-style scheduling. This is disclosed and fits the monitoring purpose, but it creates ongoing automated activity if enabled.

User impactIf you configure the cron job, the monitor may keep running and sending alerts until you remove the scheduled task.

RecommendationEnable cron only if you want continuous monitoring, and keep the documented stop command or crontab entry available for removal.

Agentic Supply Chain Vulnerabilities

SeverityLowConfidenceHighStatusNote

scripts/xiaohongshu-monitor.sh

OPENCLAW="/opt/homebrew/bin/openclaw"

The script depends on a specific local OpenClaw binary path, while the provided requirements declare no required binaries and no OS restriction. This is an under-declared dependency rather than evidence of malicious behavior.

User impactThe skill may fail or behave differently on systems without that exact CLI path, and users may not realize the dependency from metadata alone.

RecommendationDeclare the OpenClaw CLI dependency and OS/path expectations in metadata or make the binary path configurable.

Permission boundary

Checks whether tool use, credentials, dependencies, identity, account access, or inter-agent boundaries are broader than the stated purpose.

Identity and Privilege Abuse

SeverityLowConfidenceHighStatusNote

SKILL.md

首次使用需要在浏览器中登录小红书一次，之后脚本会自动使用登录态

The skill intentionally relies on a logged-in Xiaohongshu browser profile/session. This is disclosed and purpose-aligned, but it is still sensitive account/session access.

User impactThe script can access Xiaohongshu as the logged-in browser profile when it checks the monitored page.

RecommendationUse a dedicated browser profile or account if possible, and only log in if you are comfortable with the skill using that session for monitoring.

Sensitive data protection

Checks for exposed credentials, poisoned memory or context, unclear communication boundaries, or sensitive data that could leave the user's control.

Insecure Inter-Agent Communication

SeverityMediumConfidenceHighStatusConcern

scripts/xiaohongshu-monitor.sh

"$OPENCLAW" message send --target "ou_24c2bc2b000e0ea7a99dea7f4f657dbc" --message "$msg"

Notifications are sent to a fixed Feishu/OpenClaw recipient ID rather than a user-configured target, making the recipient and data boundary unclear.

User impactNew-post alerts could be delivered to an unintended Feishu account, revealing what account you monitor and the scraped post titles.

RecommendationMake the notification recipient configurable by the installing user, disclose the target clearly, and require confirmation before scheduled notifications are enabled.

Memory and Context Poisoning

SeverityInfoConfidenceHighStatusNote

scripts/xiaohongshu-monitor.sh

> "$WORKSPACE/memory/$snapshot_file"

The script stores scraped post snapshots in the OpenClaw workspace memory directory for later comparison. This is purpose-aligned but leaves persistent local records.

User impactLocal files may retain a history of monitored post titles and logs.

RecommendationReview or clear the workspace memory files if you no longer want this monitoring history retained.