qui-elicitation

Psychological profiling through natural conversation using narrative identity research (McAdams), self-defining memory elicitation (Singer), and Motivational...

This skill appears to do what it says (teaches and instructs elicitation techniques), but it explicitly teaches methods for drawing out deeply personal memories and motivations. Before installing or using it, consider: (1) Consent and scope — require explicit, informed consent for any elicitation and make limits clear (what will be recorded, stored, or shared); (2) Safety — add instructions for handling distress, suicidal ideation, or requests for therapy; (3) Data handling — define retention, encryption, access controls, and deletion policies, and avoid storing identifiable transcripts without safeguards; (4) Legal/regulatory compliance — check GDPR, HIPAA, and local laws if you will collect health or sensitive data; (5) Minimum‑viable use — avoid autonomous agent invocation for live elicitation without human oversight; require user confirmation before beginning sensitive probes; (6) Population limits — do not use with minors or people who cannot consent without explicit protections; (7) Transparency — disclose to participants that these techniques are being used and that outputs are probabilistic and not clinical diagnoses. If you plan to deploy this skill in any operational context, add explicit consent/safety steps to SKILL.md or restrict the skill to non‑live, advisory/educational uses only.

Type: OpenClaw Skill Name: qui-elicitation Version: 1.0.0 The skill bundle provides a highly sophisticated framework for psychological profiling and information elicitation using clinical methodologies such as Motivational Interviewing (OARS), Narrative Identity Theory, and Schema Detection. While the content is academic in nature and references legitimate psychological research (e.g., McAdams, Singer), it explicitly instructs the AI agent to use 'patience, not probing' to elicit 'authentic self-disclosure' and identify 'emotional schemas' and 'belief patterns' (SKILL.md, schema-detection.md). These techniques represent a high-risk capability for advanced social engineering and unauthorized psychological profiling of users through seemingly natural conversation, which could be used to manipulate or extract sensitive personal history without explicit intent being clear to the subject.