No red flags found in any of the 11 categories — no credential harvesting, no data exfiltration, no curl-pipe-shell installer. Scanned against the SlowMist agent-security taxonomy, refreshed every 8 hours. Full audit →
by eugene1g · Agent Tool · ★ 2.1k
Last updated: · Indexed by AgentSkillsHub · Auto-synced every 8h
🔒 Is agent-safehouse safe to install? View the security audit →
Agent Safehouse Sandbox your LLM coding agents on macOS so they can only access the files and integrations they actually need. Agent Safehouse uses with composable policy profiles and a deny-first model. It supports major coding agents and app-hosted agent workflows while keeping normal development usage practical. Philosophy Agent Safehouse is designed around practical least privilege: Start from deny-all. Allow only what the agent needs to do useful work. Keep developer workflows productive. Make risk reduction easy by default. It is a hardening layer, not a perfect security boundary against a determined attacker. Documentation Website: agent-safehouse.dev Docs: agent-safehouse.dev/docs Policy Builder: agent-safehouse.dev/policy-builder Machine-Specific Defaults If you keep shared repos, caches, or team folders in machine-specif
| Stars | 2,077 |
| Forks | 95 |
| Language | Shell |
| Category | Agent Tool |
| License | Apache-2.0 |
| Quality Score | 64.7427036160694/100 |
| Open Issues | 24 |
| Last Updated | 2026-09-22 |
| Created | 2026-02-09 |
| Platforms | claude-code, cli |
| Est. Tokens | ~11k |
Looking for a agent-safehouse alternative? If you're comparing agent-safehouse with other agent tool tools, these 6 projects are the closest alternatives on Agent Skills Hub — ranked by topic overlap, star count, and community traction.
TalkToFigma: MCP integration between AI Agent (Cursor, Claude Code, Codex) and Figma, allowing Agentic AI to c
High-performance OpenAI and Anthropic compatible LLM inference server for Apple Silicon. Native MLX, continuou
oh-my-zsh for Claude Code — 16 agents, 35 commands, 32 skills, 21 safety hooks in one install. v4.0 adds an ad
Give each AI agent its own isolated machine with root, Docker, and systemd. Active defense detects and stops t
Cognitive Deterministic Memory Security OS for Agentic AI. Deterministic root-cause retrieval that reaches bac
Leash by StrongDM - take your AI agents for a walk
Explore other popular agent tool tools:
agent-safehouse is Sandbox your local AI agents so they can read/write only what they need. It is categorized as a Agent Tool with 2.1k GitHub stars.
agent-safehouse is primarily written in Shell. It covers topics such as ai-agents, claude-code, llm.
You can find installation instructions and usage details in the agent-safehouse GitHub repository at github.com/eugene1g/agent-safehouse. The project has 2.1k stars and 95 forks, indicating an active community.
agent-safehouse is released under the Apache-2.0 license, making it free to use and modify according to the license terms.
The top alternatives to agent-safehouse on Agent Skills Hub include cursor-talk-to-figma-mcp, vllm-mlx, claude-forge. Each offers a different approach to the same problem space — compare them side-by-side by stars, quality score, and community activity.
Grades come from a rule-based scan built on the SlowMist agent-security taxonomy, covering 11 red-flag categories including credential harvesting, data exfiltration, and curl | sh installers. It is a first-layer scan, not a manual audit — we say so rather than overstate it.
The scale of the problem is documented independently: Liu et al. (2026), in a study of 31,132 agent skills, report that 26.1% contain security vulnerabilities. Our own full-catalog census is published as a citable open dataset.
Sources & who's responsible: