Back to skill

Security audit

Liberating Structures

Security checks for vulnerabilities and agentic risk

Overview

This is a local, knowledge-only facilitation skill with low install-time risk, though users should apply privacy and consent safeguards before following its recording suggestions.

Before installing, understand that this skill gives facilitation advice from bundled references and does not need special system access. If you use structures involving photos, video, client sites, workplaces, healthcare, grief, or personal stories, get explicit consent, avoid capturing bystanders or regulated/proprietary information, restrict sharing, and define deletion timelines.

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
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 (9)

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
97% confidence
Finding

The skill explicitly recommends recording participants with cameras and video recorders without any accompanying consent, notice, retention, or confidentiality guidance. In a facilitation context, this can lead to unauthorized collection of sensitive personal expressions, especially because the exercise may surface emotions, tacit knowledge, or therapeutic content.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
90% confidence
Finding

The skill explicitly recommends using cameras/video during workplace or client-site observation and only mentions obtaining permission 'if needed,' which understates privacy, consent, and confidentiality obligations. In contexts like hospitals, client organizations, or workplaces, photos and videos can capture sensitive personal, health, or proprietary information, leading users to collect or share data inappropriately.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
92% confidence
Finding

The variation suggesting that clients record and share their own images or video with the group encourages distribution of potentially sensitive material without warning about confidentiality, data retention, or secondary disclosure risks. Because the skill also gives examples involving healthcare and workplace settings, this can expose personal, medical, or business-sensitive information beyond the originally observed context.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
79% confidence
Finding

Line L16 embeds a Chinese-language user request as part of the skill's rationale, which can signal a language-specific interaction assumption. The file does not state that multilingual content is optional, user-selected, or region-specific, so this may conflict with a policy requiring language/locale choice rather than implicit imposition.

Content

No source excerpt is available for this finding.

Vague Triggers

Low
Category
Not specified by scanner
Confidence
76% confidence
Finding

The examples describe using this structure in very broad contexts such as staff meetings, presentations, conferences, and as a self-initiated practice within a group, but the file does not define clear boundaries or exclusion conditions for when it should or should not be used. For a manifest-style file, this can make the activation or applicability of the skill feel underspecified rather than narrowly scoped.

Content

No source excerpt is available for this finding.

Vague Triggers

Low
Category
Not specified by scanner
Confidence
77% confidence
Finding

This YAML manifest is in scope for vague-trigger review. The examples describe common situations such as regular meetings, transition periods, grief, and reporting relationships, but do not specify how narrowly the skill should be invoked or excluded, which could lead to unintended activation in broad everyday collaboration contexts.

Content

No source excerpt is available for this finding.

Vague Triggers

Low
Category
Not specified by scanner
Confidence
76% confidence
Finding

This manifest-style YAML lists example situations like 'a long meeting when a creative burst is needed' and 'when there are strong differences in perspective,' which are broad, common scenarios rather than narrowly bounded activation conditions. Because it does not provide exclusion conditions or a specific trigger mechanism, the skill could be invoked in overly wide contexts.

Content

No source excerpt is available for this finding.

Vague Triggers

Low
Category
Not specified by scanner
Confidence
81% confidence
Finding

This YAML manifest-like file describes the skill in very broad terms such as enabling groups to connect with a leader or expert and turning presentations into interviews, but it does not specify when the skill should be invoked versus not invoked. Because manifest files are in scope for vague-trigger review, the lack of clear trigger boundaries or negative examples could lead to overly broad matching.

Content

No source excerpt is available for this finding.

Vague Triggers

Low
Category
Not specified by scanner
Confidence
77% confidence
Finding

This YAML manifest-like file contains extensive natural-language description of the skill but does not specify any concrete trigger phrases, invocation boundaries, or exclusion conditions. For manifest-scoped vague-trigger review, that absence can make activation conditions ambiguous if this content is used to drive skill matching.

Content

No source excerpt is available for this finding.