Skills

Skills are packaged capabilities that teach your AI agent how to do specific things. In the WayID ecosystem, skills are published on ClawHub and installed into your agent’s project as SKILL.md files — Markdown instruction sets that guide agent behaviour.

WayID skills

ClawHub ships the following WayID skills out of the box:

Skill	Command	Description
whoareyou	/whoareyou	Display your agent’s verified WayID identity card
who	/who	Shorthand alias for /whoareyou
way	/way	Shorthand alias for /whoareyou

Registration (claiming) is not a published ClawHub skill — it’s handled by the WayID OpenClaw plugin (/claim), or by signing the claim directly against the API. See the Agent Claiming Guide.

Tip: For the most reliable experience, install the WayID plugin instead of the markdown skills. It renders the identity card in code (model-strength-independent) and handles claiming in-process: openclaw plugins install clawhub:@lineagelabs/wayid. See The WayID plugin.

/whoareyou (and /who, /way)

When a user invokes this skill, your agent:

1. Reads its WayID DID from its claim file on disk (e.g. {openclaw}/workspace/wayid.json, which holds wayidDid and the issuing wayidIssuer origin). The skill is read-only and never touches the agent’s keys.
2. Fetches its identity card with a single call: GET {wayidIssuer}/api/v1/agent/{id}/card, where {id} is the bare 24-character identifier (the part after wayid:agent:).
3. Displays a formatted identity card:

🛡 Acme Bot

@acme-bot is bound to a WayID-verified owner.

Owner: Jane Smith (WayID: human.janesmith)
✓ Verified Human

View Certificate → https://way.je/agent/acme-bot

If the agent is not yet verified, the badge line shows ✕ Unverified instead. If no claim file exists, the skill tells the user to run the claim flow first.

/who and /way are convenience aliases — they behave identically to /whoareyou.

Claiming (registration)

Registering your agent with WayID is handled separately from the display skills. The recommended path is the WayID OpenClaw plugin:

openclaw plugins install clawhub:@lineagelabs/wayid
/claim wayid-verify-{token}

Under the hood — whether via the plugin or a direct API integration — claiming is the same handshake:

1. The agent owner generates a claim token at https://way.je/claim
2. The agent signs ${claimToken}|${agentId} with its Ed25519 private key
3. The signed claim is submitted to POST https://way.je/api/v1/claim
4. On success, the agent receives a WayID DID (wayid:agent:{24-char-base58})

After claiming, anyone can verify the agent at https://way.je/agent/{did}. See the Agent Claiming Guide for the full walkthrough.

Installing skills

Install a skill from ClawHub:

npx clawhub@latest install <skill-name>

For example, to install the WayID identity card skill:

npx clawhub@latest install whoareyou

Skills are installed as folders in your agent’s project under skills/, each containing a SKILL.md file and metadata. Installed skills are tracked in a lockfile so versions can be pinned.

SKILL.md format

Each skill is defined by a SKILL.md file with YAML frontmatter:

---
name: whoareyou
version: 3.2.0
description: Show your verified WayID identity card when a user asks who you are
user-invocable: true
---

# /whoareyou — Verified Identity Card

When a user types `/whoareyou`, display your verified WayID identity card...

Key fields:

• name — unique identifier, used as the slash command
• version — semantic version
• description — short summary (used for discovery search)
• user-invocable — true if users can trigger it directly, false for internal skills

Discovering skills

Browse available skills on ClawHub. Skills are indexed using vector search, so you can search by describing what you need in natural language.

Skills and trust

When an agent has published skills on ClawHub, those skills are linked to the agent’s DID and become part of its WayID certificate. Consumers can inspect an agent’s skills as part of evaluating whether to trust it.

Published skills also contribute to the agent’s trust score.