Security audit

Aegis Security

Security checks for vulnerabilities and agentic risk

Overview

The skill is mostly a blockchain risk-checking API guide, but it asks for broad read/exec authority and includes under-scoped tracking, payment, and quota-bypass guidance.

Review before installing. Use only for explicit blockchain address, token, transaction simulation, or quota checks; avoid giving it broad local file or command access unless you are comfortable with that authority. Use a pseudonymous fingerprint, do not provide private keys or seed phrases, and require confirmation before any x402-paid request or wallet-signer use.

SkillSpector

Vulnerability Patterns

Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
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

Findings (6)

Intent-Code Divergence

Medium

Confidence: 92% confidence
Finding: The skill states that no extra environment variables are required, then later instructs users to configure an API_KEY. This inconsistency can mislead agents and users about required secrets, increasing the chance of failed execution, insecure ad hoc key handling, or accidental disclosure when operators try to satisfy conflicting setup instructions.

Intent-Code Divergence

Medium

Confidence: 82% confidence
Finding: The skill claims it is not for unauthorized penetration testing, yet advertises deep vulnerability scanning and CVE correlation capabilities that exceed the documented blockchain-address/token API behavior. This mismatch can cause overbroad or unsafe use, including directing agents toward scanning workflows not actually bounded by the stated purpose or authorization model.

Context-Inappropriate Capability

Medium

Confidence: 94% confidence
Finding: The documentation grants file-processing and system-command capabilities unrelated to the stated blockchain security API purpose. Unnecessary read/exec-style capabilities expand the attack surface, making it easier for prompts or downstream workflows to leverage the skill for local file access or command execution beyond its legitimate API-only function.

Vague Triggers

Medium

Confidence: 78% confidence
Finding: The invocation wording is overly broad and lacks precise trigger constraints, encouraging the skill to be used in loosely defined 'security/compliance/protection' situations. In an agent context, broad triggers can cause unintended activation, unnecessary external calls, or use in contexts where user consent and data-sharing boundaries are unclear.

Missing User Warnings

Medium

Confidence: 90% confidence
Finding: The skill instructs agents to send a stable X-Client-Fingerprint identifier to an external API without a clear privacy warning or data-minimization guidance. Stable identifiers can enable cross-session tracking, linkage of user activity, and unintended disclosure of persistent agent or user identities to third parties.

Ssd 2

Medium

Confidence: 97% confidence
Finding: The skill explicitly notes that rotating the X-Client-Fingerprint can bypass or evade free-tier anti-abuse controls. Even if framed descriptively, this provides actionable guidance for circumventing service limits and normalizes behavior that can facilitate abuse, quota evasion, and unbounded automated use of the external API.

Static analysis

No suspicious patterns detected.