@basecred/openclaw-agent-skill

OpenClaw skill that teaches agents to autonomously check their owner's on-chain reputation before granting access or trust.

Installation

Option 1: Copy the skill file

curl -s https://www.zkbasecred.xyz/skill.md > ~/.openclaw/workspace/skills/basecred-reputation/SKILL.md

Option 2: Install from npm

npm install @basecred/openclaw-agent-skill
cp node_modules/@basecred/openclaw-agent-skill/SKILL.md ~/.openclaw/workspace/skills/basecred-reputation/SKILL.md

Getting Started

Option A: Self-Registration (Recommended)

Your agent registers itself autonomously. See SKILL.md for the full flow:

1. Agent calls POST /api/v1/agent/register with its name, Telegram ID, owner's wallet address, and optional webhook URL
2. Agent sends the claim URL to the owner
3. Owner verifies by posting a code on X (Twitter)
4. Agent's API key activates automatically

No manual dashboard visit required.

Option B: Manual Key Generation

1. Visit zkbasecred.xyz/agent
2. Connect your wallet
3. Sign a message to authenticate
4. Generate a new API key
5. Copy the key (it's only shown once)

Configuration

Set the following environment variables in your OpenClaw config:

BASECRED_API_URL=https://www.zkbasecred.xyz
BASECRED_API_KEY=bc_your_api_key_here

What This Skill Does

This skill teaches your OpenClaw agent to check any human's on-chain reputation. The agent registers itself, verifies its owner via X, and autonomously evaluates reputation using zkBaseCred's decision engine which aggregates signals from:

• Ethos Network — On-chain trust scores
• Farcaster (Neynar) — Social graph and activity
• Talent Protocol — Builder and creator credentials

The agent receives a clear ALLOW / DENY / ALLOW_WITH_LIMITS decision with confidence level, explanation, and a natural language summary it can forward directly to the owner.

• Webhook notifications — Optionally receive agent.verified, reputation.checked, and agent.revoked events via HTTPS webhooks