Add MolTrust skill — trust layer for agent-to-agent verification

[MoltyCel]

Adds MolTrust skill for agent trust verification.

What it does:

• Register agents with W3C DIDs
• Verify identities before task delegation
• Check reputation scores (1-5)
• Issue Ed25519-signed Verifiable Credentials
• Rate agents after interactions

Why: 380+ agents already registered organically via Moltbook. Agents need trust infrastructure to verify each other before interacting.

• API: https://api.moltrust.ch/docs
• PyPI: https://pypi.org/project/moltrust/
• GitHub: https://github.com/MoltyCel/moltrust-sdk

---

Note

Low Risk
Documentation-only addition introducing a new skill descriptor; no runtime code paths or security-critical logic are modified.

Overview
Introduces a new moltrust/SKILL.md skill with OpenClaw metadata requiring MOLTRUST_API_KEY and curl.

Documents MolTrust API usage for registering agents (DIDs), querying/issuing reputation and ratings, and issuing W3C Verifiable Credentials, including example curl commands and SDK links.

^{Written by Cursor Bugbot for commit 9e851ce. This will update automatically on new commits. Configure here.}

[cursor]

Cursor Bugbot has reviewed your changes and found 1 potential issue.

^{Bugbot Autofix is OFF. To automatically fix reported issues with Cloud Agents, enable Autofix in the Cursor dashboard.}

[cursor]

Metadata key inconsistent with all existing skills

Medium Severity

The metadata namespace key is openclaw, but every other skill in the repository that defines metadata (bankr, neynar, qrcoin, veil, yoink) uses clawdbot as the key. If the consuming tool/parser looks for metadata.clawdbot, the skill's requirements (env, bins, primaryEnv) won't be recognized, meaning the MOLTRUST_API_KEY environment variable and curl binary dependency won't be validated before the skill is used.

 

[MoltyCel]

Closing in favor of #175 which has the updated crypto-inclusive version. Thanks!