Hacker News
Supe is a new open-source project presented on Hacker News that provides AI agents with a cognitive memory system, featuring validation gates and proof-of-work for tamper-evident logging, aiming to offer an audit layer for AI operations.
AI 生成摘要
Supe是一個在Hacker News上展示的新開源專案，它為AI代理提供了一個認知記憶系統，具備驗證閘和工作量證明功能，以實現防篡改的日誌記錄，旨在為AI操作提供一個審計層。
We read every piece of feedback, and take your input very seriously.
To see all available qualifiers, see our documentation.
Cognitive memory for AI agents with validation and proof-of-work
Your AI agent just mass-deleted files. Can you prove it wasn't supposed to?
Supe is the missing audit layer for AI agents. Validation gates block dangerous operations, proof-of-work creates tamper-evident logs, and persistent memory lets you query what your agent actually did.
Gates run before (pre) and after (post) every tool execution:
Every execution gets a SHA256 proof that's tamper-evident:
Query past executions with keyword search and neural spreading activation:
Output:
See CONTRIBUTING.md. Areas we'd love help with:
MIT - see LICENSE
Links: PyPI · GitHub · Issues
Cognitive memory for AI agents with validation and proof-of-work
There was an error while loading. Please reload this page.
0
2
