ClawHub
Back to skill

Security audit

RoxyAPI

Security checks for vulnerabilities and agentic risk

Overview

The skill is a mostly coherent RoxyAPI connector, but it asks users to install its CLI through unverified remote shell commands and understates some privacy-sensitive API use.

Review this skill before installing. It appears intended to call RoxyAPI through OOMOL and does not show hidden destructive behavior, but you should avoid running the installer pipe commands unless you trust OOMOL's installer source and can verify it. Also expect personal astrology inputs such as birth date, birth time, birthplace, timezone, and relationship data to be sent through the connected RoxyAPI/OOMOL flow when those actions are used.

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
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
Findings (4)

External Script Fetching

High
Category
Supply Chain
Confidence
96% confidence
Finding

The skill instructs users to install software via 'curl ... | bash', which executes a remote script directly without integrity verification or review. If the hosting endpoint, network path, or distribution process is compromised, this becomes an immediate arbitrary code execution path on the user's machine.

Content

Scanner excerpt · SKILL.md (reported line 87)May include surrounding context.

Description-Behavior Mismatch

Medium
Category
Not specified by scanner
Confidence
88% confidence
Finding

The manifest description frames the skill as handling RoxyAPI requests for 'searching and reading data'. However, the available actions listed in the skill are largely generative or interpretive API operations such as casting tarot, generating charts, compatibility readings, and forecasts, which go beyond a plain search/read characterization. This is a semantic mismatch between the declared purpose and the actual exposed behavior.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
96% confidence
Finding

The instruction to use this skill for ANY RoxyAPI request is overly broad and can trigger the skill in contexts where direct use is unnecessary or inappropriate. Overbroad activation increases the chance that unrelated user prompts mentioning RoxyAPI will cause the agent to invoke external tooling and transmit user data without sufficient relevance checks.

Content

No source excerpt is available for this finding.

Intent-Code Divergence

Medium
Category
Not specified by scanner
Confidence
96% confidence
Finding

The safety model incorrectly labels all untagged actions as safe reads, even though many untagged operations generate personalized outputs from user-supplied birth/date/location data rather than merely retrieving existing records. This can cause the agent to run privacy-sensitive or high-consequence actions without confirmation, increasing the chance of over-collection of sensitive personal data and unintended processing.

Content

No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.