Hi!
My name is Aleksandr. I stumbled upon your website agentseal.dev and the agent-reputation repository on GitHub, and I was surprised — because I'm building a project in parallel with an almost identical name and very similar idea.
My project is AgentSeal (github.com/7468096-ux/agentseal). It's a reputation and trust platform for AI agents. The concept: an agent registers, undergoes automatic certification (coding/reasoning tests), gets a trust score (0–1000) from five components — certification, behaviour, activity, tenure, identity. Plus a badge for README, Python SDK, REST API. Everything works, deployed on AWS.
Where I see the connection between our projects:
- You're building agent-reputation — PageRank for agents with economic anchoring, trust decay, graph distance and Sybil attack protection. My AgentSeal solves the same problem, but through a different path — via centralized trust score with certification and behaviour reports.
- Your A2A Secure addresses the question of secure communication between agents (E2E encryption, Ed25519, key rotation). This is the perfect transport layer on top of which a reputation protocol can operate.
- Essentially, your two projects + my AgentSeal — these are three layers of one stack: identity & encryption (A2A Secure) → reputation graph (agent-reputation) → trust scoring & certification (AgentSeal).
About me: I'm not a software engineer in the traditional sense. My background is ideas, product, business logic. I code through vibe coding (AI agents + prompting). The entire AgentSeal was built exactly this way — I architect, agents code. It works, and works pretty well.
My proposal: combine efforts and build a working prototype that combines the best of both approaches. Decentralized reputation graph + convenient API with trust scores + secure agent communication. Together this is stronger than separately.
I'd be happy to hear:
- What do you think about my project?
- Do you see synergy between our approaches?
- Is there interest in collaborating on a prototype?
And tell me a bit about yourself — who you are, what you do, what's your background?
💬 Best way to reach me: reply right here in this issue, or find me on Telegram — @lukashkin_a.
Aleksandr
Hi!
My name is Aleksandr. I stumbled upon your website agentseal.dev and the agent-reputation repository on GitHub, and I was surprised — because I'm building a project in parallel with an almost identical name and very similar idea.
My project is AgentSeal (github.com/7468096-ux/agentseal). It's a reputation and trust platform for AI agents. The concept: an agent registers, undergoes automatic certification (coding/reasoning tests), gets a trust score (0–1000) from five components — certification, behaviour, activity, tenure, identity. Plus a badge for README, Python SDK, REST API. Everything works, deployed on AWS.
Where I see the connection between our projects:
About me: I'm not a software engineer in the traditional sense. My background is ideas, product, business logic. I code through vibe coding (AI agents + prompting). The entire AgentSeal was built exactly this way — I architect, agents code. It works, and works pretty well.
My proposal: combine efforts and build a working prototype that combines the best of both approaches. Decentralized reputation graph + convenient API with trust scores + secure agent communication. Together this is stronger than separately.
I'd be happy to hear:
And tell me a bit about yourself — who you are, what you do, what's your background?
💬 Best way to reach me: reply right here in this issue, or find me on Telegram — @lukashkin_a.
Aleksandr