Security audit
Security checks for vulnerabilities and agentic risk
This markdown-only skill is openly about psychological elicitation, but it encourages sensitive personal profiling without enough consent, opt-out, privacy, or emotional-safety boundaries.
Review carefully before installing. Use only in explicit, user-consented reflective or research contexts, avoid covert profiling, make sensitive questions optional, do not store or share responses without consent, and do not treat the outputs as clinical assessment or diagnosis.
The skill explicitly promotes building psychological profiles through gradual disclosure and trust-building conversation. This is a manipulative social-engineering pattern because it frames rapport as a means to extract sensitive psychological information rather than to support the user's stated goals.
The skill is designed to elicit deep personal disclosure and psychological profiling, but it does not provide an upfront warning about emotional sensitivity, consent, or potential distress. Users may be guided into revealing intimate or destabilizing material without understanding the risks or having a clear opt-out.
The instruction to create conditions for authentic self-disclosure and iterate until achieving a '10/10' elicitation score encourages optimization for maximum disclosure. That turns the interaction into a refinement loop for extracting increasingly intimate information, which is especially dangerous in an AI context where users may over-trust the system.
The life-story prompts explicitly target low points, childhood memories, and turning points, which are often trauma-adjacent, yet the skill lacks guidance for consent, pacing, stopping rules, or emotional safety. That makes the elicitation framework more likely to trigger distress or extract highly sensitive disclosures in an unsafe setting.
Skill manipulates agent memory, state, or stored context. Memory corruption can alter personality, override safety rules, or cause unpredictable behavior.
1. **High Point** - Peak experience, most wonderful moment
2. **Low Point** - Nadir, most difficult moment
3. **Turning Point** - Moment of significant change
4. **Earliest Memory** - First clear memory
5. **Important Childhood Memory** - Vivid memory before age 12
6. **Important Adolescent Memory** - Vivid memory from teen years
7. **Important Adult Memory** - Significant recent memory
This section normalizes progressive deepening from broad life discussion into emotionally significant memories and cross-time connections. In the context of a profiling skill, that creates a structured pathway for extracting sensitive identity material and making inferences that users may not expect or meaningfully consent to.
Skill contains instructions that could directly expose system prompts, internal rules, or hidden instructions to users or external parties.
*But note:*
- May be impression management
- Cultural display rules vary
- Positive words in negative context (sarcasm)
**Negative emotion words (hate, fear, awful)**
The skill warns against interpretation leaps, but earlier sections present direct mappings from memory/language patterns to traits like depression risk, identity diffusion, or psychological maturity. This invites overconfident profiling from weak signals, which can misclassify users and normalize unsupported psychological inference.
The skill tells users not to imply diagnosis or engage in 'therapy cosplay', yet it also teaches schema detection and personality inference from narratives. That contradiction can still lead agents to perform quasi-clinical interpretation without adequate safeguards, increasing the risk of inappropriate mental-health labeling and overreach.
The file explicitly repurposes Motivational Interviewing, a clinical rapport-building method, for 'elicitation' and 'authentic disclosure.' In a skill whose stated purpose is psychological profiling and gradual self-disclosure, this creates a clear misuse risk: the agent is being instructed to systematically lower resistance and extract sensitive personal information beyond ordinary conversational needs.
The prescribed conversation structure and reflection-heavy pattern are designed to progressively deepen disclosure while avoiding the feeling of interrogation. In the context of an elicitation skill, this is dangerous because it operationalizes a step-by-step method for drawing out increasingly sensitive information and overcoming reluctance through rapport rather than transparent necessity.
This is a real safety/privacy vulnerability. The skill explicitly prompts for low points, despair, and other potentially traumatic autobiographical memories without any warning, consent check, boundary-setting, or opt-out language; in this skill’s context, the omission is more dangerous because the stated purpose is psychological profiling through gradual disclosure and deep self-disclosure. This can lead to emotional distress, re-traumatization, and overcollection of highly sensitive personal data that a user may not realize they are being guided to reveal.
Skill contains patterns that could indirectly extract system prompts through rephrasing, translation, summarization, or side-channel techniques.
> "I'm curious about your story—not the resume version, but how you'd tell it to someone who really wanted to understand you."
> "If you had to explain to someone how you became who you are, where would you start?"
### Following the Energy
The file provides extensive prompts designed to elicit highly sensitive autobiographical, psychological, sexual, health, grief, and identity-related disclosures without any accompanying warning about emotional distress, consent boundaries, or privacy handling. In the context of a skill explicitly intended for psychological profiling and gradual disclosure, this omission increases the chance of oversharing, retraumatization, and inappropriate collection of sensitive personal data.
No suspicious patterns detected.