Fuushyn Armour
fuushyn-armour is a Claude Code plugin for the Ship phase that acts as a firewall for AI agents with policy, proxy, and recovery-oriented controls.
by fuushyn · github.com/fuushyn/armour
Install a firewall layer so Claude Code agents cannot blast past your tool and network policies.
Add it to Claude Code
Install the plugin in Claude Code. One command, paste-ready.
/plugin install fuushyn-armour@fuushyn/armourBuilt to be called by your agent
Skillselion is itself an MCP server. Your agent can pull this entry and a paste-ready install config straight from the API - no copy-paste.
Retrieve this entry with skillselion.get_details("plugin:fuushyn/armour") and the paste-ready config with skillselion.get_install_config("plugin:fuushyn/armour").
What it does
fuushyn-armour is a Claude Code plugin bundle described as a firewall for AI agents. Solo and indie builders who ship agent workflows often need something between the model and the outside world: not full enterprise zero-trust, but explicit policy, proxy-style gating, controls, and recovery when an agent oversteps. Catalogue metadata points at armour, controls, policy, proxy, recovery, and security-enhanced tools—suggesting you register it when Claude Code can invoke shells, HTTP, or third-party integrations and you want guardrails before scale. It is a plugin marketplace install, not a single SKILL.md ritual; expect configuration aligned with your threat model rather than automatic safe defaults. Pair it with secret scanning and code review skills rather than treating it as a substitute for least-privilege API keys.
Highlights
- Agent-oriented firewall with policy and control surfaces for tool and proxy traffic
- Keywords cover proxy, policy, controls, recovery, and security-enhanced tooling
- Single-plugin bundle focused on defence-in-depth for autonomous coding agents
- Fits stacks where one compromised tool call could exfiltrate secrets or hit arbitrary URLs
- Community catalog entry (fuushyn/armour) with 1 bundled plugin
Why builders use it
Autonomous agents can call tools and networks with the same power as your terminal, but most solo setups have no policy layer between the model and production side effects.
After you register armour, agent traffic can be constrained by firewall-style rules so policy violations are blocked or recovered instead of silently executing.
At a glance
- Type - Plugin in Security.
- Adoption - 0 installs, 13 stars, 0 votes.
FAQ
Who is fuushyn-armour for?
Solo builders and small teams using Claude Code with agent tools who need firewall-style policy and proxy controls rather than trusting the model’s judgment alone.
When should I use fuushyn-armour?
Use it when you are wiring integrations or shipping agent automation and want controls, recovery, and policy enforcement before granting broad network or shell access.
How do I add fuushyn-armour to my agent?
Add the fuushyn/armour Claude Code plugin from the marketplace to your Claude Code environment, then configure policy and proxy settings per the plugin README in that repository.
Comments
Share how you use fuushyn-armour, gotchas, or tips for other indie builders.
No comments yet - be the first to share how you use it.