A domain-anchored cryptographic identity protocol for AI agents and verifiable credentials integrated with Openclaw Skills.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install agentpin
2. Manual Installation
Copy the skill folder to one of these locations
~/.openclaw/skills/ <project>/skills/ Priority: Workspace > Local > Bundled
3. Prompt Installation
Copy this prompt to OpenClaw to install it automatically.
Help me install agentpin using Clawhub. If Clawhub is not installed, install it first (npm i -g clawhub).
Prefer to download?
Get the raw skill files in a ZIP archive.
What is AgentPin?
AgentPin is a sophisticated cryptographic identity protocol designed specifically for the AI agent ecosystem. As a core component of the trust stack in Openclaw Skills, it enables organizations to establish domain-anchored identities for their agents. By publishing identity documents at well-known endpoints and utilizing ES256-signed JWTs, AgentPin ensures that agent interactions are verifiable, short-lived, and resistant to common identity attacks.
This skill is essential for developers building multi-agent systems where trust and provenance are critical. It bridges the gap between raw AI capabilities and secure enterprise requirements by providing a standardized way to issue, rotate, and verify agent credentials across different environments.
AgentPin Use Cases
- Establishing a root of trust for AI agents using existing DNS domains.
- Issuing time-limited JWT credentials for secure inter-agent tool access.
- Preventing man-in-the-middle attacks through Trust On First Use (TOFU) key pinning.
- Implementing capability-based security where agents delegate specific permissions to sub-agents.
- Verifying agent identity in air-gapped or restricted network environments using trust bundles.
How AgentPin Works
- The organization generates an ECDSA P-256 (ES256) keypair to serve as the agent's identity root.
- Public identity metadata is published to the domain's /.well-known/agent-identity.json endpoint for discovery.
- The agent issues a short-lived JWT credential signed with its private key when interacting with other services.
- The verifying party discovers the agent's public key from the domain and performs a strict 12-step verification process.
- The verifier checks for revocation, validates domain binding, and applies TOFU pinning to ensure the key hasn't been substituted.
- If a delegation chain is present, the verifier ensures each link narrows the scope of allowed capabilities.
AgentPin Setup
To start using AgentPin within your Openclaw Skills environment, install the library for your preferred language:
# For Rust users
cargo add agentpin
# For JavaScript/Node.js environments
npm install agentpin
# For Python developers
pip install agentpin
You can also use the CLI to generate keys and test credentials:
cargo run -p agentpin-cli -- keygen --output-dir ./keys --agent-name "my-agent"
AgentPin Data Schema & Taxonomy
AgentPin utilizes a structured JSON format for discovery and verification. The primary data structures include:
| Component | Description |
|---|---|
| Discovery Document | A JSON file at /.well-known/agent-identity.json containing agent metadata and public keys. |
| JWT Credential | An ES256-signed token containing issuer, agent ID, and capabilities. |
| Trust Bundle | A JSON package combining discovery and revocation data for offline use. |
| Revocation List | A dynamic list of revoked keys or agent IDs for real-time security updates. |
AgentPin Advanced Features
- Strict ES256-only enforcement to prevent algorithm confusion and downgrade attacks.
- 128-bit nonce-based challenge-response protocol for secure mutual authentication.
- Pluggable discovery resolvers supporting HTTP .well-known, DNS TXT records, and manual configuration.
- Offline-first design with trust bundles for CI/CD pipelines and high-security zones.
- Directory listing capabilities allowing verifiers to enumerate available agents on a domain.
SKILL.md
Loading
Related Openclaw Skills
Objective Evaluator
A sophisticated auditing skill that evaluates products and designs using the classic principles of Tony Fadell, Don Norman, and Dieter Rams.
MBTI Persona
A sophisticated configuration skill that maps the 16 Myers-Briggs personality types to your AI agent to customize communication and decision-making styles.
Hypii Hyperliquid Trading Agent
An AI-powered trading agent for Hyperliquid perpetual futures featuring automated strategies and integrated micropayments.
Stable Browser CDP
A high-stability browser automation layer using Chrome DevTools Protocol to replace unreliable extension-based relays.
SchemaPin
A cryptographic security framework for signing and verifying AI tool schemas to prevent tampering and man-in-the-middle attacks.
Skill Creator
A comprehensive framework and toolkit for developing modular extensions that equip AI agents with specialized procedural knowledge and tool integrations.
Featured*
