ClawHub
Back to skill

Security audit

self-relationship

Security checks for vulnerabilities and agentic risk

Overview

This is a self-reflection conversation skill with broad topic triggers but no code, data access, hidden install behavior, or persistence beyond normal skill installation.

Before installing, be aware that the skill may activate for broad self-reflection topics. It is suitable for reflective conversations, but not a substitute for medical, psychological, crisis, legal, or other professional support.

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 (14)

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
91% confidence
Finding

The skill describes activation for broad, loosely defined topics such as self-relationship, self-acceptance, personal growth, and 'similar topics' without clear limits. In an agent skill system, this can cause the skill to auto-activate in contexts where the user did not intend a psychology-oriented framing, creating prompt-scope overreach and potentially shaping sensitive conversations inappropriately.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The auto-load description says the skill loads automatically when users mention listed topics, but it does not specify precision constraints, exclusions, or disambiguation rules. Because this skill operates in emotionally sensitive self-reflection contexts, overbroad triggering can lead to unsolicited psychological framing, misapplication in crisis scenarios, or interference with more appropriate safety or domain-specific behaviors.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
90% confidence
Finding

The README says the skill activates when users talk about several example topics 'y temas similares', but it does not clearly bound what counts as similar. This ambiguity can cause unintended invocation for ordinary self-help or reflective conversations that only loosely overlap with the listed themes.

Content

No source excerpt is available for this finding.

Skill Enumeration

Medium
Category
Agent Snooping
Confidence
85% confidence
Finding

Skill enumerates or reads other installed skills. Access to other skills' SKILL.md files or the skills directory reveals prompt instructions, capabilities, and secrets that should be invisible to peer skills.

Content

Scanner excerpt · README.es.md (reported line 32)May include surrounding context.

bash
# Para agentes que admiten skills, p. ej. Claude Code, Trae, etc.
# Copia el directorio self-relationship a tu directorio de skills
cp -r self-relationship ~/.claude/skills/

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
92% confidence
Finding

The file states the skill loads automatically when the user mentions certain themes, but it does not specify whether exact phrases are required, how close paraphrases are handled, or when the skill should not load. Without negative examples or contextual constraints, common discussion of personal growth could collide with the trigger.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
94% confidence
Finding

The README states the skill auto-loads when users mention broad phrases such as self-understanding, self-growth, or self-acceptance. These are common conversational topics, so the trigger scope may activate the skill in many ordinary contexts and unexpectedly inject its guidance into sensitive emotional discussions. In a self-reflection/psychological support skill, overbroad activation is more dangerous because users may interpret the resulting responses as authoritative support in moments of distress.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
94% confidence
Finding

The README says the skill will auto-load when users mention very broad self-reflection topics, without clear exclusions or tighter routing criteria. Broad activation can cause the skill to engage in contexts involving mental health distress, crisis, diagnosis-seeking, or unrelated discussion, increasing the chance of overreach, inappropriate guidance, or prompt-scope interference with other more suitable safeguards.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
90% confidence
Finding

The README says the skill activates when users talk about several topics 'and similar topics,' which leaves the trigger boundary undefined. Because this is a markdown file and trigger scope is described in natural language without exclusions or negative examples, the skill may match ordinary reflective conversation more broadly than intended.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
91% confidence
Finding

The automatic-loading section says the skill loads when users mention broad self-help themes, including generic concepts like self-acceptance and self-growth. In agent systems with keyword- or topic-based routing, this can cause unintended activation in sensitive conversations, potentially steering users toward non-clinical guidance when a more appropriate safety, crisis, or domain-specific skill should handle the interaction.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
92% confidence
Finding

The README says the skill activates when a user talks about 'self-relationship', 'self-acceptance', 'understanding oneself', 'personal growth' and 'similar topics'. This trigger description is broad and includes everyday conversational themes, and the phrase 'similar topics' leaves the activation boundary unclear.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The file states the skill loads automatically when users mention several broad themes, including personal growth and self-understanding, in Russian or Chinese. Because no exclusions or contextual limits are given, the skill could activate during normal conversation where the user did not intend to invoke a specialized self-reflection skill.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
91% confidence
Finding

The README states that the agent will automatically load this skill when users mention very broad topics such as self-understanding, self-growth, and related terms. In a multi-skill agent, such broad activation criteria can cause unintended invocation in ordinary conversation, increasing the chance that the skill steers sensitive emotional discussions without explicit user intent or a tighter routing check.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
92% confidence
Finding

The activation description is very broad and covers common topics like self-understanding, self-growth, and self-acceptance without clear routing limits. That can cause the skill to trigger in many ordinary conversations, including adjacent mental-health or advice-seeking contexts, increasing the chance of inappropriate takeover, policy bypass through specialization, or responses outside the intended safety boundary.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
80% confidence
Finding

Most of the skill instructions and description are written in Chinese, with the English version only appearing later in the file. There is no explicit statement that the assistant should match the user's preferred language or ask for language choice, so the skill may effectively default to Chinese behavior without opt-in.

Content

No source excerpt is available for this finding.