🧪 Test?View on arXiv
From Proposal to Verified Effect: Praxa, an Evidence-Bound Harness for Governed AI Agent Execution
Author1, Author2, Author3, Author4, Author5
governed AIevidence-based architectureagent executionverification
2610.00015
Builder Relevance
1h ago70%
Abstract
Praxa is an evidence-bound architecture that explicitly represents the transitions from proposal to verified effects in AI agent execution.
Reality Card
Core Claim
Praxa achieves an evidence-bound architecture that makes authority-to-effect transitions explicit and testable.
Method / Result
The author-run repository-local audit passed 1,027/1,027 unit tests and 89/89 Workerd tests.
Limitations
Current evidence does not establish adversarial security, production safety, or user benefit.
Paper to code
Verified implementation resources so builders can test the paper’s claims instead of stopping at the abstract.
No verified implementation link has been attached yet. AIBuzzHub will keep this panel separate from unverified search results.