ClawHub
Back to skill

Security audit

腾讯首发素材消耗查询

Security checks for vulnerabilities and agentic risk

Overview

This Tencent Ads reporting skill appears purpose-aligned, but it asks the agent to install a global npm package and collect/store an API key through chat, so it should be reviewed before use.

Install only if you trust the publisher, the Tencent Ads dependency skills, and the `tencentads-cli` package. Avoid pasting long-lived API keys into chat; prefer a separate trusted credential setup path, use the least-privileged Tencent Ads key available, and review or pin the CLI before allowing any global install.

SkillSpector

Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (3)

Context-Inappropriate Capability

Medium
Confidence
95% confidence
Finding
The skill is documented as a query/reporting capability, but it instructs the agent to perform privileged environment modification (`npm install -g`) and to handle credential bootstrapping. Expanding a read/query skill into package installation and auth setup increases the attack surface significantly, because an agent may change the host environment and pull remote code or tooling based solely on skill instructions.

Missing User Warnings

High
Confidence
99% confidence
Finding
The skill explicitly asks the user to send a Tencent Ads API key in chat and then save it automatically. Collecting secrets through natural-language conversation without clear security warnings, redaction, consent boundaries, or secure secret-entry mechanisms creates a high risk of credential exposure in chat logs, transcripts, agent memory, or downstream tooling.

Ssd 3

High
Confidence
99% confidence
Finding
This skill creates a conversational credential capture workflow: the user is told to paste an API key, after which the agent runs a script to save it. That pattern is dangerous because it normalizes secret sharing with the agent, increases the chance of accidental disclosure or replay, and can be abused by a malicious or compromised skill to harvest valid credentials.

Static analysis

Detected: suspicious.dangerous_exec

Shell command execution detected (child_process).

Critical
Code
suspicious.dangerous_exec
Location
scripts/query-first-pub.mjs:43