diff --git a/README.md b/README.md
index 3dcaaecf4..acbe644da 100644
--- a/README.md
+++ b/README.md
@@ -3983,6 +3983,7 @@ Tools for conducting research, surveys, interviews, and data collection.
 - [aurelio-nakamura/cmdxray](https://github.com/aurelio-nakamura/cmdxray) [![aurelio-nakamura/cmdxray MCP server](https://glama.ai/mcp/servers/aurelio-nakamura/cmdxray/badges/score.svg)](https://glama.ai/mcp/servers/aurelio-nakamura/cmdxray) 📇 🏠 🍎 🪟 🐧 - Offline safety gate for agent shell commands: rates each command as danger, caution or none and explains every flag, pipe, redirect and subshell.
 - [Top-Celestial-Company-Ltd/DROS-VajraClaw-Hacker](https://github.com/Top-Celestial-Company-Ltd/DROS-VajraClaw-Hacker) [![Top-Celestial-Company-Ltd/DROS-VajraClaw-Hacker MCP server](https://glama.ai/mcp/servers/Top-Celestial-Company-Ltd/DROS-VajraClaw-Hacker/badges/score.svg)](https://glama.ai/mcp/servers/Top-Celestial-Company-Ltd/DROS-VajraClaw-Hacker) 🐍 🏠 🐧 🪟 🍎 - Deterministic execution governance gateway and W3C DID security guardrail for AI agent MCP tool calls.
 - [Jimil-Joshi/blastradius-mcp](https://github.com/Jimil-Joshi/blastradius-mcp) [![Jimil-Joshi/blastradius-mcp MCP server](https://glama.ai/mcp/servers/Jimil-Joshi/blastradius-mcp/badges/score.svg)](https://glama.ai/mcp/servers/Jimil-Joshi/blastradius-mcp) 📇 🏠 🍎 🪟 🐧 - Zero-trust security proxy, pre-execution blast radius simulation, DLP secret redaction, and cryptographic audit logging for MCP.
+- [verax-ai/verax](https://github.com/verax-ai/verax) [![verax-ai/verax MCP server](https://glama.ai/mcp/servers/verax-ai/verax/badges/score.svg)](https://glama.ai/mcp/servers/verax-ai/verax) 📇 🏠 🪟 🐧 - Sits between an agent and its tools: every call passes a policy gate and leaves a signed decision record before it runs, a refusal is recorded like an approval, payments wait for an operator on the same machine, and the ledger stays there and verifies offline. Apache-2.0.
 
 ### 🌐 <a name="social-media"></a>Social Media