ClawHub
Back to skill

Security audit

uibc-team-rob

Security checks for vulnerabilities and agentic risk

Overview

This skill is a small UIBC judging aid with a local checker and no evidence of hidden access, persistence, exfiltration, or destructive behavior.

Install this if you want a Chinese-language UIBC/ROB criteria helper. Use it for explicit TH-ROB-001 or UIBC embodied-AI migration/accountability judging tasks, and avoid relying on its broad trigger keywords for unrelated robotics or governance questions.

A.I.G

Vulnerability Patterns
  • Skill Instruction HijackingAlters the agent's session goals or safety constraints when the skill loads
  • Agent Memory PoisoningWrites attacker-controlled rules into memory that affect later sessions
  • Remote Payload Retrieval and ExecutionFetches external code whose behavior can change after review
  • Embedded Malicious CodeShips malicious scripts inside the skill and executes them locally
  • Unauthorized Access and Privilege EscalationObtains permissions beyond the task's legitimate needs

SkillSpector

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 (4)

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
89% confidence
Finding

The trigger phrase list is very broad and does not define clear scope boundaries, exclusions, or disambiguation rules. In an agentic system, this can cause the skill to activate in unrelated contexts involving generic terms like memory migration, evidence packages, or responsibility attribution, leading the agent to follow the wrong workflow or apply domain-specific governance logic where it does not belong.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The skill metadata description and read_when guidance define very broad invocation triggers, including generic terms like UIBC, embodied AI, evidence packages, accountability, and team references. This can cause the skill to be selected in conversations that only tangentially mention those topics, leading to unintended use of domain-specific judgment criteria and potentially inappropriate authoritative-sounding outputs.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
93% confidence
Finding

The '何时用' section repeats ambiguous activation guidance and expands scope to broad trigger words without clear boundaries or disambiguation rules. In an agentic system, this increases the chance of accidental invocation, which may bias downstream analysis toward this skill's rigid domain framing or cause it to answer outside its appropriate context.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The file’s docstring, usage text, and user-facing output are entirely in Chinese, and the script hardcodes Chinese messages without offering a language selection or documenting a locale-specific constraint. This is a natural-language policy concern because it imposes a specific language on users without opt-in.

Content

No source excerpt is available for this finding.