Show HN: Soul Protocol – an open standard for portable AI identity
3 コメント
Author here. I built this because I kept rebuilding the same agent personality every time I switched frameworks. Discord bot one week, Slack bot the next, Claude Code after that. Each time I'd lose everything the agent had learned.
The psychology stack came from frustration with RAG-based memory. Vector similarity treats a casual "nice weather" the same as "my mom passed away." Human memory doesn't work that way. We model significance gating (LIDA), emotional salience (Damasio), and activation decay (ACT-R) so the agent remembers what actually matters.
Quick start if you want to try it:
Happy to answer questions about the memory architecture, the eval methodology, or the spec design.
The psychology stack came from frustration with RAG-based memory. Vector similarity treats a casual "nice weather" the same as "my mom passed away." Human memory doesn't work that way. We model significance gating (LIDA), emotional salience (Damasio), and activation decay (ACT-R) so the agent remembers what actually matters.
Quick start if you want to try it:
pip install soul-protocol
soul init "MyAgent"
soul observe "I love building open source tools"
soul recall "what do I enjoy"
soul status
The .soul file is just a ZIP. Rename it, read the JSON inside. No magic.Happy to answer questions about the memory architecture, the eval methodology, or the spec design.
Problem: Your AI agent is locked into your platform. Want to deploy it elsewhere? You rebuild from scratch.
Solution: Export your agent as a .soul file (ZIP archive with JSON). It contains personality, memory, bonds, and skills. Deploy on any platform. Switch frameworks. Keep your soul.
What's in it: - Portable: .soul files work on any platform that implements the protocol - Psychology-informed: Damasio somatic markers, ACT-R activation decay, LIDA significance gating, Klein self-model - Empirically validated: Soul 8.5 vs Mem0 6.0 in head-to-head benchmarks, 20/20 judge decisions favored Soul - Zero LLM dependency: Heuristic engine works offline - CLI integration: `soul inject claude-code` writes identity + memory into any agent's config in ~50ms (6 platforms) - Multi-soul MCP: Load and switch between multiple souls in a single session - Open standard: Like HTTP or MCP. Reference implementation in Python, TypeScript next
We published RFCs for community feedback. The spec is open for input.
GitHub: https://github.com/qbtrix/soul-protocol Whitepaper: https://soul.qbtrix.com/whitepaper.html Landing page: https://soul.qbtrix.com
Would your framework benefit from portable agents? What's blocking adoption?