ClawHub
Back to skill

Security audit

software_test_case_design

Security checks across malware telemetry and agentic risk

Overview

This is a markdown-only test-case design skill with a broad Agent trigger, but no hidden execution, credential handling, persistence, or data exfiltration behavior was found.

This skill is reasonable to install if you want help drafting test cases, including API and AI agent test cases. Be aware that the word "Agent" alone may activate it unexpectedly, so users or maintainers should prefer more specific phrasing such as "AI agent testing" or "agent test case design."

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 (3)

Vague Triggers

Medium
Confidence
86% confidence
Finding
The skill description is extremely broad and covers many common testing-related requests, which increases the chance that the skill is invoked outside its intended scope. Over-broad activation can cause instruction collisions, incorrect file loading, or unintended behavior in downstream agent workflows, especially when used in multi-skill routing systems.

Vague Triggers

High
Confidence
97% confidence
Finding
Using the standalone keyword 'Agent' as a trigger is dangerously generic because it appears in many unrelated contexts, including AI discussions, software architecture, support tooling, and ordinary prose. In an agentic environment, this can cause the skill to activate on unrelated requests and override more appropriate skills or instructions, creating misrouting and potentially unsafe task handling.

Vague Triggers

High
Confidence
96% confidence
Finding
The instruction mapping table repeats the same unconstrained 'Agent' trigger, reinforcing unsafe routing behavior throughout the skill. Because the table formalizes loading of agent-testing capabilities based on that broad match, unrelated prompts containing 'Agent' could systematically pull in the wrong capability set and produce inappropriate outputs.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

Detected: suspicious.exposed_secret_literal

File appears to expose a hardcoded API secret or token.

Critical
Code
suspicious.exposed_secret_literal
Location
references/examples/format-spec.md:104