Back to skill

Security audit

Seedphrase

Security checks for vulnerabilities and agentic risk

Overview

The skill does not show data theft or persistence, but it is labeled for seed phrase work while delivering generic finance-style reference text without seed phrase safety warnings.

Review this skill carefully before installing. Do not enter or paste any real seed phrase into it. It appears to be a static placeholder-style reference tool with no observed exfiltration or persistence, but its seedphrase framing is under-scoped for a domain where mistakes can cause irreversible wallet loss.

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
  • 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
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (6)

Tp4

High
Category
MCP Tool Poisoning
Confidence
98% confidence
Finding

The skill is presented as a seedphrase reference tool for highly sensitive blockchain recovery material, but the documented behavior is a generic brochure-like command set unrelated to mnemonic handling. This mismatch can mislead users into trusting the skill in a high-risk context, increasing the chance they disclose or rely on it for seed phrase tasks without receiving appropriate safeguards or relevant guidance.

Content

No source excerpt is available for this finding.

Description-Behavior Mismatch

Medium
Category
Not specified by scanner
Confidence
97% confidence
Finding

The description and commands indicate a broad blockchain/finance reference tool rather than a seedphrase-specific utility, which is unsafe because users may invoke it in a context involving wallet recovery secrets. In security-sensitive workflows, ambiguous or misleading scope can cause operators to paste sensitive mnemonic data into an inappropriate tool or rely on irrelevant guidance.

Content

No source excerpt is available for this finding.

Intent-Code Divergence

Medium
Category
Not specified by scanner
Confidence
94% confidence
Finding

The documentation claims seedphrase applicability, but the command descriptions focus on generic topics like regulations, instruments, and strategies that are unrelated to mnemonic recovery security. This inconsistency is dangerous because it creates false expectations in a domain where misuse can directly lead to wallet compromise or irreversible asset loss.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
94% confidence
Finding

A skill operating in a seedphrase context should prominently warn that seed phrases are extremely sensitive secrets and must never be shared, logged, or sent to online tools. The absence of any warning materially increases the risk that users expose recovery phrases to the agent or associated scripts, potentially resulting in total loss of cryptocurrency assets.

Content

No source excerpt is available for this finding.

Intent-Code Divergence

Medium
Category
Not specified by scanner
Confidence
93% confidence
Finding

The inline documentation states this is a seedphrase tool, but the implemented outputs do not provide seedphrase-focused functionality. Because seed phrases protect blockchain assets directly, misleading labeling in this context is more dangerous than ordinary documentation drift: users may rely on the tool for sensitive operational decisions and receive irrelevant or incomplete guidance.

Content

No source excerpt is available for this finding.

Description-Behavior Mismatch

Medium
Category
Not specified by scanner
Confidence
96% confidence
Finding

The skill is advertised as a seedphrase reference tool for blockchain contexts, but the help text and command set provide largely generic finance, compliance, and operations content instead of seedphrase-specific guidance. In a security-sensitive context, this mismatch can mislead users or downstream agents into trusting the tool for wallet recovery or key-management decisions, causing unsafe handling of highly sensitive seed material.

Content

No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.