Security audit

is-bullshit

Security checks across malware telemetry and agentic risk

Overview

This skill is a disclosed, default-off fact-checking helper, with the main caution that it can review and summarize prior chat context when invoked or enabled.

Safe to install as an optional fact-checking aid. Leave automatic mode off unless you want fact-check blocks after every response, and avoid invoking it in chats where summarizing earlier sensitive context would be undesirable.

SkillSpector

By NVIDIA

Vulnerability Patterns

Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
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 (2)

Vague Triggers

Medium

Confidence: 97% confidence
Finding: The trigger list includes generic phrases like "verify," "check," and "are you sure," which are common in ordinary conversation and can cause the skill to activate when the user did not intend a full conversation-wide audit. Because the skill inspects the entire conversation history and emits a fact-check block, accidental activation can expose prior prompts, responses, or tool-usage context beyond the user's immediate request.

Vague Triggers

Medium

Confidence: 95% confidence
Finding: When enabled, the skill is configured to run after every response automatically, which creates a persistent post-response behavior that may override normal conversational boundaries and repeatedly analyze all prior turns. This broad auto-activation increases the risk of unintended disclosure of sensitive context and can degrade agent safety by causing unrequested transformations of every answer.

VirusTotal

61/61 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.