ClawHub
Back to skill

Security audit

aggclaw

Security checks across malware telemetry and agentic risk

Overview

This skill is a disclosed AppGrowing/YouCloud ad creative analysis integration that sends user prompts to its API and can download returned creative assets when requested.

Install only if you are comfortable sending creative-analysis prompts and related session context to AppGrowing/YouCloud using your API key. Avoid entering confidential campaign plans, client secrets, or personal data, and review downloaded creative files before reuse.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (3)

Vague Triggers

Medium
Confidence
90% confidence
Finding
The trigger list includes very broad natural-language phrases such as 'analyze creatives', 'creative analysis', and 'global campaigns', which can plausibly appear in ordinary conversation and unintentionally invoke the skill. Because invocation causes external API calls using a configured credential, accidental activation can lead to unintended data transmission, quota consumption, and user confusion.

Vague Triggers

Medium
Confidence
88% confidence
Finding
The keyword-based auto-detect behavior lacks precise activation boundaries and combines semantic classification with trigger routing, increasing the chance that unrelated user text is treated as a skill invocation. In this skill, that ambiguity is more dangerous because it can autonomously choose a mode and send the user's original input to a third-party service without an explicit command.

Autonomous Decision Making

Medium
Category
Excessive Agency
Content
1. **Check API Key** (above)
2. **Detect language + determine chat_mode** (above)
3. **Call the API directly** — do not ask the user for product positioning, target market, or other supplementary info; let the API search on its own:
   - New analysis → send `input` + `chat_mode` + `language_code`, no `session_id`
   - Follow-up → reuse `session_id`, omit `chat_mode`/`language_code` (preserve the session's mode & language)
   - Output `output`; translate if its language differs from the user's; save `session_id` for follow-ups
Confidence
84% confidence
Finding
The instruction to 'Call the API directly' and not ask the user for supplementary information authorizes autonomous external action based on incomplete or ambiguous prompts. In this context, the risk is amplified because the skill sends user-provided content to a remote service and persists session state, so a mistaken classification or prompt injection-like phrasing could trigger unintended third-party disclosure or unnecessary API usage.

VirusTotal

62/62 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.