ClawHub
Back to skill

Security audit

TRPG Data vs Context Hooks

Security checks for vulnerabilities and agentic risk

Overview

This is a plain Markdown guide for reorganizing game briefing files, with the main caution that it may prompt file moves, deletions, and broad natural-language hook keywords.

Install this only if you work with this TRPG briefing_package layout. Review any proposed file moves, deletions, and config.yaml hook keywords before applying them, especially if your project uses common words that could trigger extra context unexpectedly.

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
  • 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 (3)

Natural-Language Policy Violations

High
Category
Not specified by scanner
Confidence
97% confidence
Finding

The entire skill body is written in Chinese and presents prescriptive operational instructions, with no indication that language should follow the user's preference. If invoked in a different-language session, it can override user expectations, reduce operator comprehension, and cause misconfiguration or missed security-relevant details because the agent may follow instructions the user cannot easily audit.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
92% confidence
Finding

The skill metadata advertises broad trigger phrases such as 'reorganize data files', 'data vs context hooks', and 'context_hooks', which are generic enough to match many benign repository or packaging tasks. That can cause the skill to activate in contexts where the user did not intend it, leading to unsolicited file reorganization guidance or automated changes.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The guidance explicitly recommends using natural-language keywords without prefixes and says players should say them 'naturally', which encourages unconstrained hook triggers. In systems that inject content based on keyword matches, this increases accidental activation, over-injection of context, and possible prompt-surface expansion whenever ordinary conversation includes those words.

Content

No source excerpt is available for this finding.