Grant Thinking General

• Outcome: Pass
• Latest audit: May 29, 2026 at 8:36 PM UTC
• Version:
  1.0.0

Overview

This is a text-only grant proposal reasoning skill with no code execution, credentials, external tools, or hidden data handling.

Install this if you want an agent to automatically apply grant fundability and reviewer-risk analysis to research proposal work. Prefer project-level installation or disable implicit invocation if you only want it used when explicitly requested.

SkillSpector

By NVIDIA

Vulnerability Patterns

• Trigger Abuse: Overly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
• Prompt Injection: Instruction Override, Hidden Instructions, Exfiltration Commands
• Data Exfiltration: External Transmission, Env Variable Harvesting, File System Enumeration
• Privilege Escalation: Excessive Permissions, Sudo/Root Execution, Credential Access
• Supply Chain: Unpinned Dependencies, External Script Fetching, Obfuscated Code

Findings (1)

Vague Triggers

Medium

• Confidence: 90% confidence
• Finding: The skill enables implicit invocation without any visible trigger constraints, exclusions, or contextual guardrails. That increases the chance the agent will auto-apply grant-evaluation behavior in unrelated conversations, which can cause unintended data exposure to the skill, inappropriate steering of responses, or prompt-scope confusion if the skill is invoked in contexts outside proposal analysis.

VirusTotal

66/66 vendors flagged this skill as clean.