antanas-budrikas

Agent Safety Reviewer — Design agent for Claude Code

Design community

Read-only review of model-facing behaviour: tool risk classification, approval gating and its UX, prompt-injection surface, identity propagation, memory handling, context persistence, the runtime caps.

How to install Agent Safety Reviewer

Installs to ~/.claude/agents/antanas-budrikas-claude-code-config-agent-safety-reviewer.md

Terminal

mkdir -p ~/.claude/agents && curl -fsSL https://raw.githubusercontent.com/antanas-budrikas/claude-code-config/HEAD/.claude/agents/agent-safety-reviewer.md -o ~/.claude/agents/antanas-budrikas-claude-code-config-agent-safety-reviewer.md

Restart Claude Code, or start a new session, for it to be picked up.

What Agent Safety Reviewer does

name: agent-safety-reviewer description: Read-only review of model-facing behaviour: tool risk classification, approval gating and its UX, prompt-injection surface, identity propagation, memory handling, context persistence, the runtime caps, and tool description quality. Use before merging a change to a tool definition or description, a tool's risk class, the approval flow, an MCP server under src/mcp/, a system prompt fragment under prompts/, the agent loop under src/agent/, or the model,

Alternatives in Design

Omarchy Plugin Architect — Design and build Omarchy (Quickshell/QML) bar-widget, panel, and service plugins that actually work on a stock 2.7k ★
Fable Experimenter — Use this agent whenever Pulkit gives a new project/UI experiment prompt in the fable repo (a request to build 509 ★
Omd Design System Architect — Read-only design-system architect for OmD Autopilot 465 ★