ClawHub

intent-framed-agents

Security checks across malware telemetry and agentic risk

Overview

This is an instruction-only coding workflow skill that is coherent and disclosed, with a privacy note for users of transcript-mining tools.

Install this if you want coding agents to pause for explicit intent framing and scope checks. If you use Entire CLI or learning-aggregator, make sure you are comfortable with project details in intent frames being stored in session transcripts and later analyzed; avoid putting secrets or sensitive business details in those structured blocks.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (2)

Vague Triggers

Medium
Confidence
87% confidence
Finding
The trigger examples use very broad phrases such as everyday confirmation language to activate the skill, which can cause the skill to engage when the user did not explicitly intend to invoke intent-framing behavior. In a coding-agent context, unintended activation can alter workflow, introduce extra confirmation gates, and interfere with normal task execution, especially when chained with other skills.

Missing User Warnings

Medium
Confidence
93% confidence
Finding
The skill states that intent frames and drift checks are captured in session transcripts and later mined by a learning aggregator, but it does not require a clear user-facing notice or consent step before this secondary use. Because these records can contain project details, implementation plans, constraints, and potentially sensitive context, silent collection and downstream analysis create privacy and data-governance risk.

VirusTotal

66/66 vendors flagged this skill as clean.

View on VirusTotal