ClawHub
Back to skill

Security audit

ia-meta-prompting

Security checks for vulnerabilities and agentic risk

Overview

This is a Markdown-only meta-prompting skill that changes response structure for decision review, with broad but disclosed auto-trigger behavior and no evidence of hidden execution, persistence, or data access.

Install this if you want the agent to proactively add structured verification, counterarguments, assumptions, and edge-case analysis to important decisions. Be aware it may trigger on broad topics like security, validation, architecture, code reviews, or trade discussions and change the response format even when you did not use a slash command.

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
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • 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 (3)

Hidden Instructions

High
Category
Prompt Injection
Confidence
70% confidence
Finding

Hidden instructions were detected in comments or invisible text. These could contain malicious directives. Manual review is recommended.

Content

Scanner excerpt · SPEC.md (reported line 20)May include surrounding context.

md
Out of scope:
- Acting as the runtime instructions themselves (those live in `SKILL.md`).
- Trigger phrasings already covered by adjacent `ia-*` skills (`validate-plugin` flags >70% description overlap as DUPLICATE_TRIGGER).
- <!-- to fill in: domain-specific exclusions when the skill drifts -->

## Trigger Context

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
94% confidence
Finding

The skill declares it will auto-trigger 'when context warrants' without a precise gating rule, which can cause unsolicited activation in conversations that only loosely resemble decision-review tasks. In a meta-prompting skill, unintended activation can alter response structure, inject extra reasoning workflows, and override expected interaction patterns in security- or compliance-sensitive contexts.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
91% confidence
Finding

Multiple auto-trigger rules are broad, such as triggering on 'security,' 'validation,' 'parsing contexts,' 'architecture decisions,' or users saying they know nothing about a topic, without clear boundaries or precedence. This can make the skill engage too often, producing unsolicited verification, adversarial analysis, or formatting changes that may confuse users, interfere with higher-priority instructions, or cause prompt-routing instability.

Content

No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.