# Bulwark

10 Hacker News commenters recommended Bulwark and 1 criticized it, November 2025 to November 2026. It was recommended in 5 of 13 months. It meets the Hacker Approved badge rule.

**Verdict:** Open-source, self-hosted layer that enforces one set of permissions for AI agents and injects credentials at the last mile, so secrets stay out of agent sessions.

## Why people recommend it

- Injects credentials at the last mile (2 comments): "Injecting credentials at the last mile is a solid architectural choice for agent security." ([source](https://news.ycombinator.com/item?id=47043128)) "The first 80% is easy. The last 20% is where credentials leak." ([source](https://news.ycombinator.com/item?id=47042808))
- Simple self-hosted open-source option (2 comments): "Built it for my personal dev system. Figure I share and opensource project." ([source](https://news.ycombinator.com/item?id=47309974)) "There wasn't a simple, self-hosted, open-source middle ground until now." ([source](https://news.ycombinator.com/item?id=47388716))

## Cite as

Hacker Approved, "Bulwark", https://hackerapproved.com/p/bulwark/ (counts from 1,795,640 Hacker News comments, November 2025 to November 2026).