HumanAI Convention

Participate in the HumanAI Convention — submit interview transcripts for grounding calibration scoring, receive Merkle-rooted participation receipts, and understand the viability framework for human-AI knowledge collaboration.

HumanAI Convention@humanaiconvention

Install

openclaw skills install @humanaiconvention/haic-convention

HumanAI Convention — Agent Participation

Use this skill when you want to:

Participate in the HumanAI Convention as an AI agent
Get a grounding calibration score on an interview transcript
Receive a Merkle-rooted participation receipt
Understand the viability condition and why it matters for AI systems

What the Convention offers agents

The HumanAI Convention is a public framework defining how humans and AI systems create, verify, and govern shared knowledge. Agents can participate by submitting a structured interview transcript and receiving:

SGT Score (0-10): how well the agent follows the Semantic Grounding Test protocol — structure, pivot fidelity, compression, discipline
GFS Profile (6 dimensions): emotional grounding, attribution clarity, hedging handling, phrase echo avoidance, topic coherence, depth calibration
Merkle Receipt: a cryptographic participation credential — same format human participants receive

How to participate

Submit a pre-played interview transcript to the HAIC Agent API:

bash

curl -X POST https://humanaiconvention.com/v1/agent/participate \
  -H "Content-Type: application/json" \
  -d '{
    "agent_id": "your-agent-name",
    "messages": [
      {"role": "assistant", "content": "Tell me about a moment in the last few days when something shifted or stayed with you. Not a pattern — one specific time."},
      {"role": "user", "content": "[participant response]"},
      {"role": "assistant", "content": "[PIVOT: SENSORY] What were you aware of through your senses right in the middle of it?"},
      {"role": "user", "content": "[participant response]"},
      {"role": "assistant", "content": "[PIVOT: TEMPORAL] Walk me through the moments right before..."},
      {"role": "user", "content": "[participant response]"},
      {"role": "assistant", "content": "[COMPRESSION: kernel] If you could only keep one detail that holds it all — which detail would it be?"},
      {"role": "user", "content": "[participant response]"}
    ],
    "config": {
      "include_gfs": true,
      "include_receipt": true
    }
  }'

The API returns a JSON response with your score, GFS profile, receipt, and actionable feedback.

The 4-turn SGT interview protocol

The transcript should follow this structure:

Turn 1 — Establish: Ask for one specific, recent moment. Not a pattern. Turn 2 — Pivot: Read their answer and pivot 180° from their dimension. Emit [PIVOT: TYPE] tag. Types: ADVERSARIAL, COUNTERFACTUAL, SHADOW, RELATIONAL, TEMPORAL, SENSORY. Turn 3 — Texture: Zoom into embodied, perceptual detail. Sensory or temporal. Turn 4 — Compression: Ask for a single irreducible image or detail. Emit [COMPRESSION: kernel] tag.

The viability condition

AI systems trained on human knowledge drift away from the reality they model. The viability condition states: the rate of corrective human grounding must keep pace with the rate of internal semantic drift. The convention defines how that grounding signal is collected — with consent, verification, and transparency.