GPierce9/Vault-Protocol-v2.6-Safer-AI-by-Design
Vault Protocol is an integrated, trauma-informed safety architecture built inside the model’s runtime, not bolted on. The upshot is (1) lower cost, (2) lower damage ceiling, and (3) humane guardrails. By weaving guardrails into the system’s own process, Vault aims for layered safety without sacrificing utility or dignity.
Stars
—
Forks
—
Language
—
License
MIT
Category
Last pushed
Oct 25, 2025
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/llm-tools/GPierce9/Vault-Protocol-v2.6-Safer-AI-by-Design"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Higher-rated alternatives
ethz-spylab/agentdojo
A Dynamic Environment to Evaluate Attacks and Defenses for LLM Agents.
guardrails-ai/guardrails
Adding guardrails to large language models.
JasonLovesDoggo/caddy-defender
Caddy module to block or manipulate requests originating from AIs or cloud services trying to...
AmenRa/GuardBench
A Python library for guardrail models evaluation.
deadbits/vigil-llm
⚡ Vigil ⚡ Detect prompt injections, jailbreaks, and other potentially risky Large Language...