Tendermint

Security checks across malware telemetry and agentic risk

Overview

This is a low-risk static reference script, but it is mislabeled as Tendermint while mostly returning generic finance and compliance text.

Installing this does not appear to grant meaningful system or account access. Treat it as a shallow static reference only, and do not rely on it for real Tendermint, validator, compliance, operational, or investment decisions without checking authoritative sources.

SkillSpector

By NVIDIA

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 (5)

Tp4

High

Category: MCP Tool Poisoning
Confidence: 95% confidence
Finding: The skill is presented as a Tendermint blockchain reference tool, but the documented capabilities are largely generic finance/compliance topics such as ROI, Sharpe ratio, regulations, and strategies. This mismatch can mislead agents or users into invoking the skill in security- or finance-sensitive contexts under false assumptions, increasing the risk of inappropriate guidance, bad decisions, or prompt-routing abuse.

Description-Behavior Mismatch

Medium

Confidence: 91% confidence
Finding: The command set is dominated by generic finance/reference subjects rather than Tendermint-specific blockchain operations or documentation. In an agent ecosystem, this can cause incorrect tool selection and trust transference, where a blockchain-labeled skill supplies unrelated financial content that may be acted on without proper scrutiny.

Intent-Code Divergence

Medium

Confidence: 94% confidence
Finding: Claiming Tendermint-specific functionality while documenting unrelated finance/compliance behavior is a deceptive interface contract, even if unintentional. This makes the skill more dangerous in context because Tendermint/blockchain workflows are operationally sensitive, and inaccurate tooling can misdirect incident response, compliance interpretation, or production decisions.

Description-Behavior Mismatch

Medium

Confidence: 95% confidence
Finding: The skill is presented as a Tendermint reference tool, but the exposed commands and content are generic finance, compliance, and investment-style material rather than Tendermint-specific guidance. This semantic mismatch can mislead users and downstream agents into trusting irrelevant or incorrect output in blockchain workflows, creating integrity and decision-making risk even though there is no direct code-execution primitive here.

Intent-Code Divergence

Medium

Confidence: 94% confidence
Finding: The inline documentation explicitly brands the script as a Tendermint tool, while the implementation largely returns unrelated generic reference text. In an agent setting, this kind of deceptive labeling increases the chance that users or automated systems will invoke the tool for blockchain tasks and act on materially wrong guidance, making the mismatch a real trust and safety issue.

VirusTotal

66/66 vendors flagged this skill as clean.

View on VirusTotal