Top 13 Rust ai-safety Projects
-
template-repo
Agent orchestration & security template featuring MCP tool building, agent2agent workflows, mechanistic interpretability on sleeper agents, and agent integration via CLI wrappers (by AndrewAltimit)
Project mention: Template for secure AI multi-agent coding workflows | news.ycombinator.com | 2026-02-06 -
SaaSHub
SaaSHub - Software Alternatives and Reviews. SaaSHub helps you find the best software and product alternatives
-
Rullst
📜🦀🌐 Full-Stack Web Framework for Rust 🌐🦀📜"Rust for those who want to build securely and easily, but not suffer"🤖 The Next-Gen Paradigm: Designed for Humans & AI. Rullst is built from the ground up to be a developer-first framework, optimized for seamless AI-assisted coding and effortless human productivity.
-
phantom-secrets
Stop AI coding agents from leaking your API keys. Local proxy + MCP that swaps real secrets for phm_ tokens — works with Claude Code, Cursor, Windsurf, and Codex.
Project mention: Show HN: Phantom – Let AI use your API keys without leaking them | news.ycombinator.com | 2026-03-28 -
execwall
OpenClaw Execution Firewall - Seccomp-locked AI agent sandbox with policy-enforced command governance, WhatsApp/Telegram integration, and Python isolation. Deploy on Oracle Cloud Free Tier.
Project mention: Execwall – firewall to stop ModelScope CVE-2026-2256 (AI agent command injectn) | news.ycombinator.com | 2026-03-13GitHub: https://github.com/sundarsub/execwall
-
veto
🔑 Human-in-the-loop confirmation for AI Agents. Let your AI deploy, delete, push — with your fingerprint. (by runkids)
Project mention: Show HN: Veto – Intercept dangerous commands before AI executes them | news.ycombinator.com | 2026-01-27- Full audit trail
Currently supports Claude Code (via hooks). Looking into Moltbot integration but it doesn't expose hooks yet — if anyone has ideas, I'm all ears.
Feedback welcome.
https://github.com/runkids/veto
-
vellaveto
Agentic security control plane for MCP and AI agent tool calls. MCP-native policy gateway with topology discovery and audit. (by paolovella)
Project mention: Show HN: VellaVeto – fail-closed gateway for MCP tool calls | news.ycombinator.com | 2026-03-28 -
shield
Local guardrails for AI coding agents. Wraps any MCP server and blocks destructive tool calls — DROP TABLE, rm -rf, force-push, unscoped UPDATE/DELETE — before they execute. Free, open-source, runs entirely on your machine. (by AperionAI)
Project mention: Aperion Shield v0.7 – guardrails for AI coding agents now run as Git hooks | news.ycombinator.com | 2026-05-26 -
project-guardian
AI Guardian Firewall — a local, user-space, agent-agnostic firewall that mediates an autonomous AI agent's actions (files, shell, network, services) with a deterministic policy boundary, a tamper-evident audit log, and a human-in-the-loop approval cockpit. No kernel modules. Apache-2.0.
Project mention: A user-space firewall that gates an AI agent's actions | news.ycombinator.com | 2026-06-29 -
fishnet
Security proxy for AI agents (OpenClaw, LangChain, CrewAI). Protects API keys, enforces spend limits, blocks dangerous endpoints. Local-first Rust binary — your credentials never leave your machine. (by fishnetproxy)
Project mention: How I Saved 20,000 Gas Per Transaction by Reordering One Line in Solidity | dev.to | 2026-03-01While building a smart wallet contract for Fishnet — an AI agent transaction security proxy — I ran a self-imposed code review and found a subtle optimization that every Solidity developer should know about.
-
actra
Actra - control what runs before it runs, controls what actions are allowed before they execute. Evaluate policies across APIs, workflows and AI agents in real time. (by getactra)
Project mention: The Claude Code Leak Is a Warning: AI Infrastructure Is Outpacing Control | dev.to | 2026-04-05Learn more: https://actra.dev
-
aegis-shellguard
Aegis — the shell guard for AI coding agents. Destructive commands from Claude Code, Codex, or Cursor pause for your approval, with a snapshot taken first — so even a "yes" is recoverable.
Project mention: I built Aegis — a firewall between AI coding agents and your shell | dev.to | 2026-07-09I want to be honest here, because security tools that oversell themselves are a plague: Aegis is a heuristic guardrail, not a sandbox. It is not a privilege boundary. A determined attacker — or a sufficiently creative agent — can construct commands that evade pattern-based classification. There's a full threat model (https://github.com/IliasAlmerekov/aegis/blob/main/docs/threat-model.md) in the repo that spells out exactly what's in and out of scope.
-
Project mention: Show HN: HELmR – A runtime control layer for autonomous agents | news.ycombinator.com | 2026-03-09
-
Project mention: GVM-light weight governance runtime for AI Agents:Rust proxy, Linux kernel features(seccomp-bpf, namespace, overlayfs) | dev.to | 2026-04-10
View on GitHub
Rust ai-safety discussion
Rust ai-safety related posts
Index
What are some of the best open-source ai-safety projects in Rust? This list will help you:
| # | Project | Stars |
|---|---|---|
| 1 | template-repo | 130 |
| 2 | Rullst | 16 |
| 3 | phantom-secrets | 13 |
| 4 | execwall | 10 |
| 5 | veto | 7 |
| 6 | vellaveto | 6 |
| 7 | shield | 2 |
| 8 | project-guardian | 2 |
| 9 | fishnet | 2 |
| 10 | actra | 2 |
| 11 | aegis-shellguard | 2 |
| 12 | helmr-core | 0 |
| 13 | Analemma-GVM | 0 |