AgentSign: AI 에이전트를 위한 제로 트러스트 ID 및 서명
hackernews
|
|
📰 뉴스
#ai 딜
#ai 보안
#ai 에이전트
#eu ai act
#openai
#owasp
#제로 트러스트
원문 출처: hackernews · Genesis Park에서 요약 및 분석
요약
AI 에이전트의 보안 취약성과 EU AI 법규 등을 해결하기 위해 개발된 AgentSign은 모든 행동에 서명하고 신원을 검증하는 제로 트러스트 보안 솔루션을 제공합니다. 조사 결과 LangChain과 AutoGPT 등 인기 있는 12개 프레임워크는 암호화 신원이나 서명 기능이 결여되어 있어 익명의 프로세스가 운영 환경에 접근하는 위험이 있습니다. 반면 AgentSign은 HMAC-SHA256 서명을 기반으로 한 부정할 수 없는 디지털 여권을 발급하여, 어떤 시스템이든 오프라인으로 에이전트의 신원과 무결성을 검증하고 MCP 툴 연결을 제어할 수 있게 합니다.
본문
Zero trust identity for autonomous AI agents. Every action signed. Every tool gated. Every agent verified. No identity, no trust. Agent security is not optional. OWASP Agentic Top 10 and the EU AI Act now require it. We audited 12 popular agent frameworks. None have cryptographic agent identity, execution signing, or trust scoring. | Framework | Stars | Identity | Signing | Trust Score | MCP Gate | Revocation | |---|---|---|---|---|---|---| | AutoGPT | 182K | ✗ | ✗ | ✗ | ✗ | ✗ | | LangChain / LangGraph | 100K+ | ✗ | ✗ | ✗ | ✗ | ✗ | | MCP Ecosystem | 80.7K | ✗ | ✗ | ✗ | ✗ | ✗ | | OpenHands | 64K | ✗ | ✗ | ✗ | ✗ | ✗ | | Microsoft AutoGen | 50.4K | ~ | ✗ | ✗ | ✗ | ✗ | | CrewAI | 45.6K | ✗ | ✗ | ✗ | ✗ | ✗ | | HuggingFace smolagents | 25.5K | ✗ | ✗ | ✗ | ✗ | ✗ | | OpenAI Agents SDK | 19.4K | ✗ | ✗ | ✗ | ✗ | ✗ | | Google ADK / Vertex | 15.6K | ~ | ✗ | ✗ | ✗ | ~ | | NeMo Guardrails | 5.7K | ✗ | ✗ | ✗ | ✗ | ✗ | | Amazon Bedrock | Managed | ~ | ✗ | ✗ | ✗ | ~ | | Devin / Cognition AI | Closed | ✗ | ✗ | ✗ | ✗ | ✗ | | AgentSign | OSS | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ = native support ~ = partial (cloud-locked IAM) ✗ = not available Source: GitHub, March 2026. Register, onboard your agent, verify anywhere. No OAuth, no dashboard required. Pure API. Without this, your agent is an anonymous process with production access. That violates OWASP ASI03. Every agent gets a unique identity backed by HMAC-SHA256 signatures. Register, onboard, and go. The agent carries its signed passport everywhere. A self-contained, signed JSON document the agent carries everywhere. Any system can verify it offline -- no server needed. Like SSL certificates, but for AI agents. The identity layer between agents and MCP tool servers. Before any agent calls a database, file system, or API via MCP, the Trust Gate checks identity, stage, and trust score. Any service can verify an agent's passport in one API call. Check if the signature is valid, the agent hasn't been revoked, and the passport hasn't been tampered with. Sign up, onboard your agent, verify everything. That's it. Create a free account via the console or API. Get an API key instantly. No credit card required. Register your agent with one API call. It gets a cryptographic identity and signed passport. Verify any agent's identity before granting access. Public endpoint, no auth needed. No credit card required. Upgrade when you need more agents. Need unlimited agents? Contact us for Enterprise pricing. AgentSign runs wherever you need it. Start with our cloud to test, deploy on-prem for production. Same API, same SDK. Both options use the same SDK (npm install agentsign). Switch from cloud to on-prem with one config change.
Genesis Park 편집팀이 AI를 활용하여 작성한 분석입니다. 원문은 출처 링크를 통해 확인할 수 있습니다.
공유