A cryptographic security framework for signing and verifying AI tool schemas to prevent tampering and man-in-the-middle attacks.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install schemapin
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 schemapin 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 SchemaPin?
SchemaPin is a specialized security layer designed to prevent "MCP Rug Pull" attacks by enabling developers to cryptographically sign their tool schemas using ECDSA P-256 and SHA-256. It provides a robust trust stack for AI agents, ensuring that the tools they interact with have not been tampered with since their original publication. By integrating this into Openclaw Skills, developers can offer users a verifiable guarantee of integrity through deterministic JSON serialization and robust key management.
The system relies on Trust-On-First-Use (TOFU) key pinning and RFC 8615 .well-known endpoints for public key discovery. This multi-language library supports Python, JavaScript, Go, and Rust, making it a versatile choice for any developer looking to harden their agentic workflows against malicious schema modifications and ensure secure tool execution environments.
SchemaPin Use Cases
- Preventing unauthorized modifications to Model Context Protocol (MCP) tool schemas in production.
- Verifying the identity of tool developers via public key discovery using well-known endpoints.
- Securing air-gapped or offline AI environments using pre-distributed trust bundles.
- Protecting entire skill directories from file-level tampering using the SkillSigner manifest system.
How SchemaPin Works
- The developer generates an ECDSA P-256 keypair and publishes the public key to a .well-known endpoint for discovery.
- Tool schemas are processed through a canonicalization engine to ensure deterministic JSON serialization for consistent hashing.
- The schema is signed with the developer's private key, generating a cryptographic signature that accompanies the tool.
- When an AI agent loads the tool, it fetches the public key and verifies the signature against the current schema.
- On the first successful verification, the developer's key fingerprint is pinned (TOFU) to detect and reject future key substitution attacks.
- The system continuously checks revocation lists or standalone documents to ensure the signing keys remain valid and uncompromised.
SchemaPin Setup
Install the library for your preferred language to begin securing your Openclaw Skills:
Python:
pip install schemapin
JavaScript:
npm install schemapin
Go:
go get github.com/ThirdKeyAi/schemapin/[email protected]
CLI Key Generation:
schemapin-keygen --output-dir ./keys
SchemaPin Data Schema & Taxonomy
SchemaPin utilizes a manifest-driven data structure to maintain integrity. For directory signing, it generates a .schemapin.sig file with the following taxonomy:
| Field | Description |
|---|---|
skill_hash |
The SHA-256 root hash representing the entire state of the skill folder. |
signature |
The Base64-encoded ECDSA signature of the root hash. |
signer_kid |
The fingerprint of the developer's public key. |
file_manifest |
A key-value map linking every file in the directory to its specific SHA-256 hash. |
domain |
The developer's domain used for public key discovery via .well-known paths. |
SchemaPin Advanced Features
- SkillSigner: A powerful utility to sign entire directory structures (like those found in Openclaw Skills) to ensure no file has been added, removed, or modified.
- Pluggable Discovery Resolvers: Support for HTTP, Local File, and Trust Bundle resolvers via a ChainResolver fallthrough system.
- Standalone Revocation Documents: Granular control over key lifecycles with detailed revocation reasons and timestamps.
- Tamper Detection API: Programmatic detection of specific modifications, identifying exactly which files were altered in a signed directory.
- Cross-Language Compatibility: Identical cryptographic implementation across Python, Rust, Go, and JavaScript for seamless integration in polyglot environments.
SKILL.md
Loading
Related Openclaw Skills
AgentPin
A domain-anchored cryptographic identity protocol for AI agents and verifiable credentials integrated with 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.
Skill Creator
A comprehensive framework and toolkit for developing modular extensions that equip AI agents with specialized procedural knowledge and tool integrations.
gog: Google Workspace CLI
A comprehensive command-line interface for managing Gmail, Calendar, Drive, Contacts, Sheets, and Docs via OAuth.
Featured*
