DobermanCore/Doberman-Core
Local, open-source runtime guardrail that sits between AI coding agents and their tools, giving every action a PASS/AUTH/BLOCK verdict to stop destructive or exfiltrating commands before they run.
Awesome Infra for AI › Weekly › 2026-09-21
2026-09-21
AI Safety & Guardrails
nizos/probity
Hooks into Claude Code, Codex, and GitHub Copilot CLI to enforce coding rules, such as strict TDD or a ban on destructive commands, by checking every file write and shell command before it runs.
openguardrails/openafw
Local proxy that sits in front of Anthropic, OpenAI, and Gemini API calls from coding agents, replacing secrets in outgoing requests with placeholders and restoring them in responses so raw keys never reach the model or provider.
LLM Evaluation & Testing
open-compass/opencompass
Open-source LLM evaluation platform for running standardized benchmarks across models, with configurable datasets, prompt templates, and an official public leaderboard.
xinxuxin/keystone-bench
Evaluation benchmark that tests whether chat assistants change clinical advice correctly when a decisive fact is added, removed, or contradicted, and whether physician-written grading rubrics still apply after the edit.
Model Serving Frameworks
anush008/fastembed-rs
Rust library for generating text, sparse, and image embeddings and reranking scores locally via ONNX Runtime, without a Python or GPU dependency.
Workflow Orchestration for AI
Idun-Group/idun-agent-platform
Self-hosted, open-source production wrapper that turns a LangGraph or Google ADK agent into a FastAPI service with a chat UI, admin panel, tracing, guardrails, memory persistence, MCP tool governance, and prompt management.