Add rai-guard — runtime Responsible AI compliance engine (OWASP LLM Top 10, EU AI Act, NIST AI RMF)#3158
Add rai-guard — runtime Responsible AI compliance engine (OWASP LLM Top 10, EU AI Act, NIST AI RMF)#3158ashish993 wants to merge 1 commit into
Conversation
There was a problem hiding this comment.
Pull request overview
Note
Copilot was unable to run its full agentic suite in this review.
Adds a new “Agent Skills” entry to the README’s curated list of AI/LLM development libraries.
Changes:
- Add
rai-guardto the Agent Skills list with a detailed description.
💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.
| - Agent Skills | ||
| - [django-ai-plugins](https://github.com/vintasoftware/django-ai-plugins) - Django backend agent skills for Django, DRF, Celery, and Django-specific code review. | ||
| - [graphify](https://github.com/safishamsi/graphify) - Turn any folder of code, SQL schemas, R scripts, shell scripts, docs, papers, images, or videos into a queryable knowledge graph. | ||
| - [rai-guard](https://github.com/ashish993/rai-guard) - Runtime Responsible AI compliance engine that enforces OWASP LLM Top 10, EU AI Act, and NIST AI RMF via proxy, ASGI middleware, or Python decorator. Ships checks for prompt injection, PII, toxicity, hallucination, and insecure output with async SQLite audit logging. |
|
@ashish993 Thanks for your submission. However, this project does not meet our acceptance criteria: Repository age: The repository was created 1 day ago. We require a minimum of 3 months (or 6 months for Hidden Gem submissions) to ensure project stability. GitHub stars: 0 star (minimum 100 required, or strong justification for Hidden Gem). Please see our CONTRIBUTING.md for full requirements. You're welcome to resubmit once the project has matured and gained community traction. |
What is rai-guard?
rai-guard is an open-source runtime Responsible AI compliance engine for Python.
Compliance frameworks enforced at runtime:
Three integration modes — drop it into any stack:
raiguard serveAIGuardMiddlewarefor FastAPI/Starlette@guard.protectChecks: prompt injection, PII (SSN, credit cards, API keys), toxicity, hallucination risk scoring, insecure output (SQL/XSS/shell injection)
Evidence trail: async SQLite audit log (WAL mode), HTML/JSON compliance reports, live dashboard
Why
AI and Agents > Agent Skills? It adds safety/compliance enforcement as a skill layer on top of any LLM-backed application or agent framework (LangChain, LlamaIndex, OpenAI SDK, etc.).Repo: https://github.com/ashish993/rai-guard | License: Apache 2.0 | Python: 3.10+