Tencent Cloud Rum

• Outcome: Pass
• Latest audit: May 29, 2026, 8:36 PM
• Version:
  0.1.3

Overview

This is a coherent Tencent Cloud RUM analysis skill, but it requires sensitive cloud credentials and can access detailed user telemetry.

Install only if you want an agent to query your Tencent Cloud RUM data through the disclosed MCP endpoint. Use a dedicated least-privilege Tencent Cloud key, avoid sharing or committing the generated mcporter config, rotate credentials if they may have been exposed, and avoid user-targeted log searches unless they are authorized and necessary.

SkillSpector

By NVIDIA

Vulnerability Patterns

• Data Exfiltration: External Transmission, Env Variable Harvesting, File System Enumeration
• Prompt Injection: Instruction Override, Hidden Instructions, Exfiltration Commands
• Privilege Escalation: Excessive Permissions, Sudo/Root Execution, Credential Access
• Supply Chain: Unpinned Dependencies, External Script Fetching, Obfuscated Code
• Excessive Agency: Unrestricted Tool Access, Autonomous Decision Making, Scope Creep

Findings (4)

Missing User Warnings

Medium

• Confidence: 94% confidence
• Finding: The skill instructs users to provide long-lived Tencent Cloud `SecretId` and `SecretKey` credentials directly in environment variables and MCP headers, but it does not prominently warn against exposing, logging, committing, or reusing these secrets broadly. If mishandled, these credentials could grant unauthorized access to Tencent Cloud resources and monitoring data, and long-lived keys increase blast radius compared with scoped or ephemeral credentials.

Missing User Warnings

Medium

• Confidence: 92% confidence
• Finding: The example explicitly demonstrates querying RUM logs by a specific user identifier (`uin`) without any privacy guardrails, purpose limitation, authorization checks, or masking guidance. In an observability skill, this can normalize user-targeted log retrieval and enable operators or downstream agents to access potentially sensitive per-user activity and error data beyond what is necessary for aggregate troubleshooting.

Missing User Warnings

Medium

• Confidence: 95% confidence
• Finding: The documentation instructs users to place long-lived Tencent Cloud credentials directly into MCP server headers and provides no warning about secure storage, least-privilege scoping, redaction, or risks of exposing secrets in configs, logs, or shared repositories. In an agent-skill context, this is more dangerous because users may copy the example verbatim into local config files that are often persisted, synced, or inspected by other tools.

Missing User Warnings

Medium

• Confidence: 96% confidence
• Finding: The script persists long-lived Tencent Cloud credentials (SecretId and SecretKey) into a local mcporter configuration file under the user's home directory without clearly warning that the secrets will remain stored on disk. This increases the chance of credential exposure through weak file permissions, backups, workstation compromise, or accidental sharing of the config file, and the stored credentials appear to grant direct access to the RUM service via request headers.

VirusTotal

60/60 vendors flagged this skill as clean.