What cryptographic algorithm does SchemaPin use?
SchemaPin uses ECDSA P-256 with SHA-256 for signing and verification.
@ThirdKeyAI
基本情報
The SchemaPin protocol for cryptographically signing and verifying AI agent tool schemas to prevent supply-chain attacks.
ツール一覧はまだありません
この掲載には表示できるツール一覧がまだありません。サーバーにツールがないという意味ではありません。対応機能はプロジェクトのドキュメントをご確認ください。
ツールは README から自動的に抽出されます。メンテナーは ## Tools という見出しの下に記載することで、このタブに反映できます。
SchemaPin is a cryptographic tool for schema verification designed for AI agents and MCP servers. It uses ECDSA P-256 signatures, DNS-anchored trust, and Trust-On-First-Use (TOFU) key pinning to prevent "MCP Rug Pull" attacks. It is intended for tool developers who want to sign their schemas and skill folders, and for AI agents that need to verify schemas have not been tampered with.
Install the library in your preferred language: pip install schemapin (Python), npm install schemapin (JavaScript), go install github.com/ThirdKeyAi/schemapin/go/cmd/...@latest (Go), or add schemapin = "1.3.0" to your Cargo.toml (Rust). Then generate a keypair, sign a schema using SchemaSigningWorkflow, and verify it with SchemaVerificationWorkflow.
.well-known discovery for public keys (RFC 8615).well-known, local file, trust bundle, chain)SchemaPin uses ECDSA P-256 with SHA-256 for signing and verification.
Public keys are discoverable via `.well-known/schemapin.json` on the provider's domain, following RFC 8615.
Yes, via trust bundles that allow verification without network access.
TOFU (Trust On First Use) pinning records the public key seen on first interaction and warns or blocks if a different key is presented later, preventing key substitution attacks.
v1.4.0-alpha.2 adds three optional features: signature expiration (`expires_at`) with degraded-not-failed verification, DNS TXT cross-verification at `_schemapin.{domain}`, and schema version binding (`schema_version` + `previous_hash`) for opt-in lineage chain enforcement.
Painless E2E Automation for Mobile and Web

Couples therapist Figs O'Sullivan (LMFT, 17 years, endorsed by Sue Johnson, creator of EFT) inside your AI. Ask relationship questions answered from his clinical writing, name the pattern in a fight, get a three-step rep

Model Context Protocol Servers
Hindsight: Agent Memory That Learns
A curated list of references for MLOps
コメント