ClawHub

StitchFlow

Security checks across malware telemetry and agentic risk

Overview

StitchFlow is a coherent design-generation helper, but it should be used with awareness that it can run local Stitch tooling, use a Stitch API key, inspect relevant UI context, and save outputs locally.

Install only if you intend to use StitchFlow for Stitch-based UI generation. Review the referenced stitch-starter toolkit and installer before running setup, use a scoped Stitch API key if possible, and avoid using it with confidential briefs or private code unless sharing that context through Stitch is acceptable. Clear or check the saved runs/latest-screen state when switching projects.

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
92% confidence
Finding
The top-level description is broad enough to trigger on many ordinary design or UI-related requests without clear boundaries. In agentic environments, overly permissive activation can cause the skill to run in contexts where its local file access, shell-based install path, and external API workflow are unnecessary, increasing the chance of unintended tool use or data exposure from inspected codebases.

Vague Triggers

Medium
Confidence
95% confidence
Finding
The "When to use" triggers are ambiguous and expansive, especially phrases like wanting a screen generated from a rough idea or wanting visual edits, which can match many normal product conversations. Because this skill instructs the agent to inspect existing code/UI context and use a local toolkit with saved outputs, over-activation can broaden access to local project data and cause unnecessary external processing.

Vague Triggers

Medium
Confidence
88% confidence
Finding
The default prompt and description are broad enough to match common design-related requests, while the policy explicitly allows implicit invocation. This increases the chance the skill will be triggered without a clear, narrow user opt-in, potentially causing unintended local artifact creation, HTML generation, or screenshot saving from ordinary user requests.

VirusTotal

66/66 vendors flagged this skill as clean.

View on VirusTotal