Security audit

维表智联管理

Security checks across malware telemetry and agentic risk

Overview

This is a coherent management skill, but it should be reviewed because it guides sensitive credential use and business-resource changes with some under-scoped safety guidance.

Install only if you intend to let the agent help operate Dimens resources. Before using it, require explicit confirmation for write/delete/reset/public-share actions, avoid pasting real API secrets directly into command lines or chats, store returned secrets in a secrets manager, review local profile token storage, and check team/project/resource scope before running permission or public-view commands.

SkillSpector

By NVIDIA

Vulnerability Patterns

Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code

Findings (8)

Missing User Warnings

Medium

Confidence: 93% confidence
Finding: The document correctly notes that apiSecret is returned only once, but it omits essential handling guidance for a highly sensitive credential. In a skill intended to guide real operational use, this increases the chance that users will paste secrets into chats, logs, tickets, or screenshots and then be unable to recover securely, creating avoidable secret exposure risk.

Missing User Warnings

Medium

Confidence: 97% confidence
Finding: Passing --api-secret on the command line can expose the secret through shell history, process listings, CI job logs, and audit tooling. Because this skill is operational documentation, presenting the unsafe pattern without warning can directly lead users to leak live credentials during normal troubleshooting or automation.

Missing User Warnings

Medium

Confidence: 89% confidence
Finding: The documentation describes destructive operations like disable, delete, and reset-secret without warning about outage risk, credential invalidation, or rollback limitations. In an admin-oriented skill, users may execute these actions during troubleshooting and unintentionally break integrations or production access.

Missing User Warnings

Medium

Confidence: 88% confidence
Finding: The document explicitly instructs users to pass an API secret on the CLI and states that access and refresh tokens are written to a local profile, but it provides no warnings about shell history exposure, process-list leakage, file permission hardening, or secure storage. In an agent skill context, operational docs are often turned directly into automation behavior, so omission of secret-handling guidance can lead to credential disclosure or token theft on shared or compromised hosts.

Missing User Warnings

Medium

Confidence: 93% confidence
Finding: The guide includes a `doc delete` command in the normal maintenance flow without any warning about irreversibility, backup/restore checks, or explicit user confirmation. In an operational skill for creating and maintaining business resources, this increases the chance of accidental destructive actions that can remove documentation needed for operations, auditability, or collaboration.

Missing User Warnings

Medium

Confidence: 96% confidence
Finding: The guide recommends creating a view with `--is-public true` as the default remediation for a missing view, but does not explain that this may expose project data to a broader audience than intended. In a skill focused on business tables, permissions, and project setup, normalizing public access without a privacy review can cause unintended data disclosure, especially if tables contain customer, internal, or operational information.

Vague Triggers

Medium

Confidence: 88% confidence
Finding: The phrase “帮我搭一个系统” is a very broad trigger in a skill that routes users into team/project creation and initialization flows. Because this repository appears to govern operational actions over business resources, an ambiguous activation phrase can cause the agent to enter privileged orchestration paths when the user only intended to ask for high-level advice, increasing the chance of unintended project/resource creation or over-collection of context.

Vague Triggers

Medium

Confidence: 90% confidence
Finding: The repeated broad invocation phrase reinforces a default routing rule that may over-match ordinary user requests. In the context of a management skill for creating and maintaining teams, projects, tables, permissions, and workflows, this increases the risk that loosely phrased requests are interpreted as authorization to begin operational setup steps, potentially leading to unintended side effects or privilege-sensitive actions.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

Detected: suspicious.exposed_secret_literal

File appears to expose a hardcoded API secret or token.

Critical

Code: suspicious.exposed_secret_literal
Location: references/key-auth/references/examples.md:127

File appears to expose a hardcoded API secret or token.

Critical

Code: suspicious.exposed_secret_literal
Location: references/key-auth/references/login-flow.md:136