By fuushyn
Proxy MCP servers through a security layer to block destructive tool calls, audit operations, and enforce safety policies. Backup and restore configurations, monitor proxy status and protected servers, and run pre-tool reminders via hooks.
npx claudepluginhub fuushyn/armour --plugin armour🛡️ The Firewall for AI Agents - Block dangerous commands before they execute.
curl -fsSL prehooks.ai/install.sh | bash
PreHooks intercepts tool calls from AI coding assistants and blocks dangerous operations:
rm -rf / → BLOCKEDcurl ... | bash → BLOCKEDSecurity-enhanced MCP proxy for Claude Code - Block destructive tool calls and audit all operations.
curl -fsSL https://raw.githubusercontent.com/fuushyn/armour/main/scripts/install-armour.sh | bash
This downloads a prebuilt plugin bundle, installs the marketplace locally, installs the plugin, and enables it by writing to ~/.claude/settings.json.
Restart Claude Code to load the plugin.
Overrides:
ARMOUR_RELEASE_TAG=v1.0.3
ARMOUR_RELEASE_URL=https://github.com/fuushyn/armour/releases/download/v1.0.3/armour-plugin-darwin-arm64.tar.gz
ARMOUR_INSTALL_DIR=~/.armour/armour-plugin
Use ARMOUR_MARKETPLACE_SOURCE to fall back to a Git marketplace (for example, https://github.com/fuushyn/armour.git).
Add marketplace in Claude Code:
/plugin add-marketplace fuushyn/armour
Install plugin:
/plugin install armour
Restart Claude Code to load the plugin
View MCP tools:
/mcp
That's it! All your MCP servers now route through Armour with security policies applied.
Servers are configured in ~/.armour/servers.json and automatically synced on each session start.
MIT License - see LICENSE file for details.
Made with ❤️ for AI safety
Runtime security plugin for Claude Code with balanced default hooks plus the Runwall inline MCP gateway for shell, git, MCP, secret, and exfiltration risks.
Matches all tools
Hooks run on every tool call, not just specific ones
Admin access level
Server config contains admin-level keywords
Share bugs, ideas, or general feedback.
Runtime security enforcement for AI coding agents — policy hooks, audit receipts, threat hunting, and security tools for Claude Code.
A secure runtime for Claude Code. Intercepts every tool call with policy-based allow/block/ask decisions, evasion detection, path fencing, file snapshots, and audit logging.
Command Line Agent Safety Harness. All interactions with clash policy should go through this plugin
Safety for Agents - Agent Detection & Response (ADR) for Claude Code
Runtime security for AI agents. Blocks destructive actions before execution, routes high-risk operations through human approval, and maintains an immutable audit trail. Covers OWASP MCP Top 10, ASI Top 10, and Agentic Skills Top 10.