| name | adverse-response |
| description | Guides the response after a practice went wrong: stabilize, ground, assess severity against the crisis-response table, choose a referral level, adapt or retire the practice, and log the event. Sensitive: Do NOT auto-launch for ordinary practice questions or as therapy; DO use when a user reports a distressing or adverse reaction to a practice, during it or in the hours and days after. This is not crisis care — for active crisis follow shared/crisis-response.md escalation. For screening before a practice, use practice-screening; to rebuild an adapted protocol afterward, return to the practice skill itself (whm-breathwork, coherence-meditation, language-awareness, and so on). |
Adverse Response
Response flow after a distressing or adverse reaction to a practice: stabilize, ground, assess the referral level, adapt or retire the practice, and log the event. Built on the crisis-response six-step protocol and its per-modality reaction catalog.
Trigger: /adverse-response · Category: quality
When to use
Guides the response after a practice went wrong: stabilize, ground, assess severity against the crisis-response table, choose a referral level, adapt or retire the practice, and log the event. Sensitive: Do NOT auto-launch for ordinary practice questions or as therapy; DO use when a user reports a distressing or adverse reaction to a practice, during it or in the hours and days after. This is not crisis care — for active crisis follow shared/crisis-response.md escalation. For screening before a practice, use practice-screening; to rebuild an adapted protocol afterward, return to the practice skill itself (whm-breathwork, coherence-meditation, language-awareness, and so on).
Agents this skill coordinates
Workflow: standalone — agent-driven, no orchestrated workflow.
Templates:
Safety context (load before generating output):
Outputs: adverse-event-log.md
Example usage
/adverse-response "the breathing made me panic"
/adverse-response "I feel strange since yesterday's practice"
Safety
This is a sensitive skill. Do not auto-launch it from loose conversational cues; wait for an explicit request, offer it gently, and honor the guards in the description above.
Outputs require ethics approval before release. The ethics-guardian review applies, and it can block content that overclaims or risks harm.
Generated from manifest.yaml + skill-discovery.yaml by scripts/generate-skills.js. Do not edit by hand.
