ClawHub
Back to skill

Security audit

Structured Falsification

Security checks for vulnerabilities and agentic risk

Overview

This is a prompt-only decision-analysis skill with broad activation and a fixed Chinese output format, but it contains no code, credential access, persistence, or hidden external behavior.

Install this if you want concise structured comparisons for complex decisions. Be aware it may activate on broad decision prompts and may answer in a Chinese table format; for financial topics, treat its position-sizing and investment conclusions as analysis support rather than professional advice.

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
96% confidence
Finding

The skill declares auto-trigger behavior for broad, high-level categories such as uncertain decisions and multi-factor judgment, which can activate on many normal user requests without clear boundaries. This can cause unsolicited takeover of the interaction style, suppress transparency via hidden reasoning directives, and route benign queries into a constrained decision framework that may not fit the user’s intent.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The trigger phrases include very common language like 'should I', 'which one', 'what about', 'evaluate', and 'compare', which are ubiquitous across ordinary conversation. Because the skill can auto-trigger on routine prompts, it risks over-activation, inappropriate hidden behavior changes, and reduced alignment with the user’s requested level of detail or task scope.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
97% confidence
Finding

The mandated Chinese-only output format can override the user’s language preference and reduce usability, auditability, and comprehension for non-Chinese users. In a skill that may auto-trigger broadly, this becomes more problematic because users may unexpectedly receive critical recommendations in a language they did not request.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
90% confidence
Finding

This skill content introduces Chinese-language labels and later includes market-specific instructions for Chinese A-shares, but it does not explicitly tell users that the skill is bilingual or China-market-oriented. Under the policy, forcing a specific language or locale without opt-in can be a natural-language policy issue.

Content

No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.