ClawHub
Back to skill

Security audit

Aiprox Swarm

Security checks for vulnerabilities and agentic risk

Overview

Review is warranted because the skill sends user questions, optional context, and a spend token to AIProx while its credential and data-sharing disclosures are incomplete and partly contradictory.

Install only if you are comfortable sending prediction questions, added context, and market-related notes to AIProx. Treat AIPROX_SPEND_TOKEN as a billing or access secret: keep it out of prompts, screenshots, repositories, and logs, and avoid submitting confidential trading, business, personal, or regulated information until the publisher clarifies token handling and data-retention terms.

A.I.G

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

SkillSpector

Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (5)

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The skill instructs users to submit task and optional context to https://aiprox.dev/api/orchestrate but does not clearly warn that user-provided questions, news context, and potentially sensitive market data are transmitted to a third-party external service. This creates a data exposure risk because users may include confidential or regulated information assuming the skill operates locally or transiently without external disclosure.

Content

No source excerpt is available for this finding.

External Transmission

Medium
Category
Data Exfiltration
Confidence
92% confidence
Finding

This skill sends user-supplied content and a spend token to an external network endpoint. External transmission is expected for the feature, but it still represents a genuine security/privacy risk because questions, context, and billing credentials leave the local environment and are processed by a third party.

Content

Scanner excerpt · SKILL.md (reported line 46)May include surrounding context.

Make Request — Basic

bash
curl -X POST https://aiprox.dev/api/orchestrate \
  -H "Content-Type: application/json" \
  -H "X-Spend-Token: $AIPROX_SPEND_TOKEN" \
  -d '{

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
96% confidence
Finding

The examples require an X-Spend-Token header but do not include a clear credential-handling warning. Users may expose the token through copied commands, shell history, logs, screenshots, shared terminals, or insecure environment management, which could allow unauthorized use of the token and unwanted spend.

Content

No source excerpt is available for this finding.

External Transmission

Medium
Category
Data Exfiltration
Confidence
93% confidence
Finding

The context-enabled example explicitly encourages sending richer user context to the external orchestrator, increasing the chance that sensitive information, proprietary analysis, or private data is disclosed to a third party. The risk is amplified because the example normalizes attaching free-form contextual text without any safety boundary.

Content

Scanner excerpt · SKILL.md (reported line 72)May include surrounding context.

Make Request — With Context

bash
curl -X POST https://aiprox.dev/api/orchestrate \
  -H "Content-Type: application/json" \
  -H "X-Spend-Token: $AIPROX_SPEND_TOKEN" \
  -d '{

Intent-Code Divergence

Medium
Category
Not specified by scanner
Confidence
98% confidence
Finding

The skill makes contradictory security claims: it says no authentication or API key is required, while the documented requests require an X-Spend-Token sourced from an environment variable. This can mislead users into mishandling credentials, underestimating billing/security implications, or assuming the endpoint is anonymous when it is not.

Content

No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.