Back to skill

Security audit

Search Bot

Security checks for vulnerabilities and agentic risk

Overview

This skill is a disclosed web-search connector that sends search tasks and a payment token to its advertised AIProx service.

Install only if you are comfortable sending search queries and an AIPROX_SPEND_TOKEN to AIProx, with server-side use of Brave Search and Claude/LightningProx summarization as described. Avoid using it for secrets, private documents, or sensitive personal queries unless that matches your trust and data-handling expectations.

Vulnerability Patterns
  • 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
  • Unauthorized Access and Privilege EscalationObtains permissions beyond the task's legitimate needs
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
Findings (1)

External Transmission

Medium
Category
Data Exfiltration
Content
## Make Request

```bash
curl -X POST https://aiprox.dev/api/orchestrate \
  -H "Content-Type: application/json" \
  -d '{
    "task": "latest developments in AI agent frameworks this week",
Confidence
60% confidence
Finding
Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.

Static analysis

No suspicious patterns detected.