GLACIS vs LLM Guard

GLACIS puts controls in the path of AI actions and creates signed records of covered decisions for external verification. When an action crosses configured bounds, it can be constrained, held for a person, or denied; other actions can pass through unchanged. Glacis says it operates nine production components as a managed service alongside the identity provider, cloud, and security systems a customer already uses. Records use OVERT, a published open format that implementers and reviewers can inspect, and can be checked in a browser without a Glacis account. The site says payloads remain in the customer’s environment and records contain hashes; omitting protected fields depends on deployment setup. A record check does not establish whether an action was correct, whether every event was captured, or whether a system is safe or compliant. GLACIS is available by API, web, and self-hosted deployment. Its Free account costs nothing, with no card required. The contact page describes a six-week evaluation for a workflow under review.

LLM Guard is a self-hosted security toolkit for checking interactions with large language models. It scans prompts and outputs for concerns such as prompt injection, secrets, personally identifiable information, toxicity, bias, malicious URLs, and factual consistency. It can sanitize interactions, detect harmful language, help prevent data leakage, and resist prompt injection attacks. The documentation includes examples for the OpenAI SDK, LangChain, LiteLLM, and LlamaIndex. In the OpenAI examples, prompts and responses are scanned; a streaming example also checks response chunks. LLM Guard can run as an API, and custom scanners can be added. Installation uses `pip install llm-guard` and requires Python 3.9 or higher. It is free and listed under the MIT License, with API, Linux, and self-hosted platforms. The repository was archived on July 9, 2026, is read-only, and is no longer actively developed or maintained. Production deployment assistance is directed to GitHub issues, Slack, and [email protected].