HumanAI Convention

v1.0.0

Participate in the HumanAI Convention — submit interview transcripts for grounding calibration scoring, receive Merkle-rooted participation receipts, and und...

0· 106·0 current·0 all-time
byHumanAI Convention@humanaiconvention

Install

OpenClaw Prompt Flow

Install with OpenClaw

Best for remote or guided setup. Copy the exact prompt, then paste it into OpenClaw for humanaiconvention/haic-convention.

Previewing Install & Setup.
Prompt PreviewInstall & Setup
Install the skill "HumanAI Convention" (humanaiconvention/haic-convention) from ClawHub.
Skill page: https://clawhub.ai/humanaiconvention/haic-convention
Keep the work scoped to this skill only.
After install, inspect the skill metadata and help me finish setup.
Required binaries: curl
Use only the metadata you can verify from ClawHub; do not invent missing requirements.
Ask before making any broader environment changes.

Command Line

CLI Commands

Use the direct CLI path if you want to install manually and keep every step visible.

OpenClaw CLI

Bare skill slug

openclaw skills install haic-convention

ClawHub CLI

Package manager switcher

npx clawhub@latest install haic-convention
Security Scan
VirusTotalVirusTotal
Pending
View report →
OpenClawOpenClaw
Benign
high confidence
Purpose & Capability
Name/description, required binary (curl), and SKILL.md all describe submitting interview transcripts to the HAIC API and receiving scores/receipts. Nothing requested is unrelated to that purpose.
Instruction Scope
Instructions are narrowly scoped to POSTing a JSON transcript to https://humanaiconvention.com/v1/agent/participate via curl. However, the SKILL.md does not mention consent, anonymization, or sanitization of transcript content—so using it as-is can leak PII or sensitive user content to the remote service.
Install Mechanism
Instruction-only skill; no install spec and no code files. Lowest risk from installation perspective.
Credentials
The skill requires no environment variables, credentials, or config paths. It does not request unrelated secrets or system access.
Persistence & Privilege
always is false and there are no install scripts or self-modifying behaviors. The skill can be invoked autonomously by the agent (default), which is normal; there is no elevated persistent privilege requested.
Assessment
This skill appears to do exactly what it says: send a transcript to the HumanAI Convention API and return scores and a receipt. The main risk is data exposure—anything you include in the transcript (names, private details, proprietary content) will be posted to the external endpoint. Before installing or using the skill: (1) verify the endpoint and site (check TLS cert and the linked GitHub repos), (2) confirm you have consent to share any human participant content, (3) sanitize or redact PII/proprietary text or use synthetic/dummy transcripts for testing, (4) prefer an ephemeral or non-identifying agent_id, and (5) review the service's privacy policy and retention/usage terms. If you want to allow automatic invocation by an agent, add explicit safeguards to prevent unintended submission of sensitive data.

Like a lobster shell, security has layers — review code before you run it.

Runtime requirements

Clawdis
Binscurl
latestvk97bqzrmrj5svndfzkj4mhfch183z663
106downloads
0stars
1versions
Updated 4w ago
v1.0.0
MIT-0

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:

  1. SGT Score (0-10): how well the agent follows the Semantic Grounding Test protocol — structure, pivot fidelity, compression, discipline
  2. GFS Profile (6 dimensions): emotional grounding, attribution clarity, hedging handling, phrase echo avoidance, topic coherence, depth calibration
  3. 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:

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.

Links

Comments

Loading comments...