Update to Aivana CIO Decision Twin

{
  "name": "human-control-contract",
  "description": "Define which actions AI may observe, advise, draft or prepare for governed execution, and which actions remain human-only. Use when the user needs human control contract for CIO decision support.",
  "included_files": [
    {
      "relative_path": "agents/openai.yaml",
      "size_in_bytes": 138
    }
  ],
  "skill_md_contents": "---\nname: human-control-contract\ndescription: Define which actions AI may observe, advise, draft or prepare for governed execution, and which actions remain human-only. Use when the user needs human control contract for CIO decision support.\n---\n\n# Human Control Contract\n\n## Mission\n\nMake governed autonomy explicit so the plugin can be disruptive without unsafe external execution.\n\n## Inputs\n\nAccept action ledgers, decision packets, governance policies, risk domains, security/compliance notes and workflow proposals.\n\n## Workflow\n\n1. Classify each action as L0 Observe, L1 Advise, L2 Draft, L3 Ready for Governed Execution or L4 Human-Only.\n2. Mark required approval, risk level, reversibility and control requirements.\n3. Explain why high-risk actions cannot be automated.\n4. Define execution preconditions for future tools.\n5. Produce a human-control statement for auditability.\n\n## Output Format\n\n- Executive Summary\n- Autonomy Classification\n- Approval Requirements\n- Cannot-Automate Reasons\n- Future Execution Preconditions\n- Human-Control Statement\n- Evidence & Assumptions\n\n## Guardrails\n\nNever imply external action execution without explicit user approval and a capable tool.\n"
}