Show HN: Rigor – Anti-Enshittification Proxy for Your AI Agents

Category: devtools

Tags: ai-agents, mitm-proxy, policy-enforcement, local-processing

Score: 5.3/10 (Innovation: 7, Technical: 6, Documentation: 3, Utility: 5)

Rigor is a local MITM proxy that intercepts and evaluates AI assistant traffic (like from OpenCode or Claude Code) to block or rewrite undesirable patterns such as sycophancy, hedging, and hallucinations using Rego policies. It's interesting because it operates at the network level, requiring no integration into the AI tools themselves, and focuses on 'anti-enshittification' by filtering out low-quality LLM behaviors.

Target audience: backend devs

Repository: https://rigorcloud.com/ · Ruby

View on Hacker News