ClawHub
Back to skill

Security audit

AEO Analytics Free

Security checks for vulnerabilities and agentic risk

Overview

This skill is a coherent analytics helper that sends user-supplied brand prompts to Gemini or web search and stores local scan history, with no hidden execution or persistence found.

Install only if you are comfortable sending the tracked prompts and brand queries to Gemini or web search providers. Avoid scanning secrets, personal data, regulated data, or confidential client material, and prefer safer credential handling than placing API keys in command-line URLs.

A.I.G

Vulnerability Patterns
  • Insecure Skill Coding PracticesFinds exploitable flaws such as hardcoded secrets or command injection
  • Skill Instruction HijackingAlters the agent's session goals or safety constraints when the skill loads
  • Agent Memory PoisoningWrites attacker-controlled rules into memory that affect later sessions
  • Remote Payload Retrieval and ExecutionFetches external code whose behavior can change after review
  • Embedded Malicious CodeShips malicious scripts inside the skill and executes them locally
Findings (1)

T09 · Insecure Skill Coding Practices

Warning
Location
references/gemini-grounding.md:15
Finding

Gemini API Key Exposed Through Command-Line Arguments

Content
View full analysis
Remediation
View remediation

SkillSpector

Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Memory PoisoningPersistent Context Injection, Context Window Stuffing, Memory Manipulation
  • 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)

Memory Manipulation

High
Category
Memory Poisoning
Confidence
80% confidence
Finding

Skill manipulates agent memory, state, or stored context. Memory corruption can alter personality, override safety rules, or cause unpredictable behavior.

Content

Scanner excerpt · SKILL.md (reported line 155)May include surrounding context.

md
- Data file location: `aeo-analytics/<domain>.json`
- Schema: see `references/data-schema.md`
- Each scan appends to the `scans` array — never delete history
- Prompts can be added/removed without affecting historical data
- When adding new prompts, they start with 0 scans (no backfill)

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
92% confidence
Finding

The document instructs users to send prompts to Gemini with Google Search grounding, which necessarily transmits prompt contents to external Google services and may trigger additional search queries. Without an explicit warning about third-party data disclosure and handling of potentially sensitive prompts, users could unknowingly expose proprietary, personal, or customer data.

Content

No source excerpt is available for this finding.

External Transmission

Medium
Category
Data Exfiltration
Confidence
95% confidence
Finding

The provided curl example performs an outbound request to Google's API and includes user-supplied prompt text, causing external transmission of potentially sensitive data. Because grounding invokes Google Search as a tool, the effective disclosure surface is broader than a normal model call, and the documentation does not warn users about that exposure.

Content

Scanner excerpt · references/gemini-grounding.md (reported line 15)May include surrounding context.

API Call (grounding enabled)

bash
curl -s "https://generativelanguage.googleapis.com/v1beta/models/gemini-2.0-flash:generateContent?key=${GEMINI_API_KEY}" \
  -H "Content-Type: application/json" \
  -d '{
    "contents": [{"parts": [{"text": "What is the best AI travel planner that gives real local recommendations?"}]}],

Static analysis

No suspicious patterns detected.