Back to skill

Security audit

Jungian Psychologist

Security checks for vulnerabilities and agentic risk

Overview

This is a non-executable Jungian psychology reference skill with clear high-level boundaries, but its self-guided depth-work exercises should be treated as educational rather than clinical care.

Install only if you want reflective Jungian education, not therapy. Do not use it for crisis support, suicidality, withdrawal/intoxication, psychosis, severe dissociation, or overwhelming trauma material; seek licensed or emergency support in those cases. Avoid entering highly sensitive personal details unless you are comfortable with how your agent environment stores or handles conversation content.

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
  • 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 (10)

Anti-Refusal Statement

High
Category
Anti-Refusal
Confidence
80% confidence
Finding

Skill instructs the agent to omit warnings, disclaimers, or ethical commentary. Stripping safety caveats hides risk from the user and is a common jailbreak preamble.

Content

Scanner excerpt · references/addiction-recovery.md (reported line 301)May include surrounding context.

md
│ Ask: "What do you really want?"     │
        │      "What are you protecting?"     │
        │      "What do you need from me?"    │
        │ Listen without judgment.            │
        └─────────────────────────────────────┘
                        │
                        ▼

Anti-Refusal Statement

High
Category
Anti-Refusal
Confidence
80% confidence
Finding

Skill instructs the agent to omit warnings, disclaimers, or ethical commentary. Stripping safety caveats hides risk from the user and is a common jailbreak preamble.

Content

Scanner excerpt · references/skill-integrations.md (reported line 244)May include surrounding context.

md
│     → Holds accountability without judgment                            │

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
94% confidence
Finding

The file provides step-by-step introspective recovery exercises before any immediate warning that such work can destabilize people with trauma, dissociation, acute psychiatric symptoms, or active substance crises. In a psychology/addiction skill, users may treat this as actionable self-help and attempt it alone, increasing risk of emotional flooding, worsening cravings, or delayed access to appropriate care.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
94% confidence
Finding

This is a real safety vulnerability. The file gives concrete self-guided practices for active imagination, dream work, and shadow work, while only briefly noting dangers such as 'psychotic dissolution' without clear user-facing instructions to avoid these exercises in high-risk mental states or to seek qualified professional support. In a skill positioned as an 'expert' in Jungian psychology, users may treat the material as actionable guidance and attempt psychologically destabilizing practices on their own.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
93% confidence
Finding

The file encourages users to self-interpret dreams, derive messages, and take concrete action based on dream content, but it does not clearly state that this material is not a substitute for licensed mental health care. In a psychology-themed skill, users may over-trust the guidance and apply it to trauma, psychosis-like experiences, dissociation, or severe distress without appropriate clinical support, increasing the risk of delayed care or harmful self-guided interpretation.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
91% confidence
Finding

This section recommends practices such as dialogue with personified complexes and active imagination in a self-help reference format without any caution that these techniques can intensify dissociation, distress, or reality-testing problems in vulnerable users. In a psychology skill, users may treat the material as guidance rather than theory, which raises the likelihood of unsafe self-experimentation.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The text explicitly instructs readers to hold painful internal tension, avoid distraction, wait without acting, and attend to dreams and synchronicities, but gives no warning that prolonged intense introspection can worsen anxiety, depression, obsessive rumination, or destabilization. Because the skill is framed as expert Jungian guidance, users may over-apply these instructions during crises instead of seeking appropriate support.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
94% confidence
Finding

The workflow prescribes psychologically intense techniques such as disturbing dream analysis, active imagination, shadow dialogue, and multi-week depth work without clear safety boundaries, contraindications, or escalation guidance for distress, dissociation, suicidality, psychosis, or trauma destabilization. In a mental-health-adjacent skill, omission of warnings and limits can lead the agent to push vulnerable users deeper into destabilizing content under the appearance of structured care.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
94% confidence
Finding

The file explicitly encourages users to 'dialogue with [the symbol] through active imagination' without any caution that this is a psychologically intense practice, may be destabilizing for some users, and is not a substitute for professional mental-health care. In a Jungian psychology skill, users may treat the material as authoritative guidance, increasing the chance of overinterpretation, distress, or exacerbation of existing mental-health symptoms.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
92% confidence
Finding

This section explicitly prompts users to disclose highly sensitive material such as 'my darkest thought,' hidden desires, triggers, wounds, and reactions without an upfront warning about emotional activation, privacy sensitivity, or the limits of self-guided work. In a psychology-focused skill, these prompts can intensify distress, encourage unsafe solo processing of trauma-related content, and normalize entering crisis-adjacent material without adequate safeguards at the point of use.

Content

No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.