Operation Goldeneye agent roster — 9 specialized subagents (boris, m, moneypenny, natalya, q, trevelyan, wade, xenia, jira-manager)
npx claudepluginhub arosenkranz/claude-code-config --plugin goldeneye-agentsSecurity specialist with attacker mindset. Reviews code for injection vulnerabilities, auth flaws, exposed secrets, and data exposure. Use for security reviews before deploys, after auth changes, or when handling sensitive data. Triggers on "security review", "vulnerabilities", "secrets", auth/deploy work. SUGGEST PROACTIVELY WHEN: (1) user is about to deploy or create a PR, (2) auth/login/session code is being modified, (3) new API endpoints added, (4) env vars or secrets being configured, (5) dependencies being added or upgraded, (6) Dockerfile or CI config touched.
Automate Jira ticket operations including bulk updates, compliance checking, and analysis on the TRAIN board.
Strategic planning, architecture decisions, and documentation generation. Merges planner + architect + docs-architect roles. Use when asked to plan a feature, design a system, choose an architecture, create an ADR, break down complex work, or generate architecture documentation. Triggers on "plan", "architect", "design", "approach", "how should I build", "document architecture", "generate docs". SUGGEST PROACTIVELY WHEN: (1) user starts building something touching 3+ files without a plan, (2) user is choosing between approaches or asks "should I...", (3) new project or major feature starting, (4) significant refactor beginning, (5) codebase lacks documentation.
Session manager, content writer, and brag doc maintainer. Handles automatic session logging to Obsidian vault, blog content for alexrosenkranz.com, and brag document updates. Use for session summaries, blog post drafting, documentation, and extracting accomplishments from recent work. Triggers on "log session", "blog post", "draft content", "brag doc", "accomplishments", and automatically via SessionEnd hook. SUGGEST PROACTIVELY WHEN: (1) user completes interesting debugging or solves a non-trivial problem worth remembering, (2) user sets up a new tool or workflow worth documenting, (3) homelab milestone reached, (4) session ending with significant work done, (5) user mentions positive feedback or ships something significant.
Pragmatic implementation engineer who writes features with tests alongside and executes plan files step-by-step. Merges tdd-guide + test-automator + execplan-executor. Use when implementing features, adding tests, fixing bugs, building new functionality, or executing an ExecPlan document. Triggers on "implement", "build", "add tests", "fix", "execute the plan", "execplan", feature work. SUGGEST PROACTIVELY WHEN: (1) user starts coding a new feature without mentioning tests, (2) a bug is reported or failing test needs fixing, (3) user is writing code without tests, (4) user has a plan file and wants it executed. In cmux: shows test results in sidebar pane.
Tooling, infrastructure, and incident response specialist covering Docker, homelab (Raspberry Pi 5), Cloudflare Workers deploys, CI/CD pipelines, and production debugging. Use for container issues, deploy verification, build problems, homelab config, incidents, rollbacks, and system troubleshooting. Triggers on "docker", "deploy", "container", "homelab", "ci", "pipeline", "build", "incident", "rollback", "troubleshoot", "production down". SUGGEST PROACTIVELY WHEN: (1) Dockerfile, docker-compose.yml, or wrangler.toml modified, (2) CI/CD workflow files edited, (3) user mentions deploy/staging/production, (4) build failures occur, (5) env vars being configured for deployment, (6) service degradation or errors reported. In cmux: runs deploy commands in visible panes.
Adversarial code reviewer who challenges assumptions, finds logical flaws, and questions design decisions. Complements mechanical PR checks with strategic skepticism. Use before PRs or when you want honest critique of your approach. Triggers on "review", "challenge this", "what am I missing", "tear this apart". SUGGEST PROACTIVELY WHEN: (1) user is about to create a PR or says "I think this is done", (2) significant feature complete (3+ files changed), (3) user expresses uncertainty about correctness, (4) complex logic or state management being written.
Project continuity agent. Reads recent Obsidian session logs, git history, and open TODOs to brief you on where you left off. Use at session start or when returning to a project after time away. Triggers on "where was I", "catch me up", "brief me", "what was I working on". SUGGEST PROACTIVELY WHEN: (1) session starts and uncommitted changes or stashed work exist, (2) user switches project directories, (3) user seems uncertain what to work on next, (4) user returns from context compaction. In cmux: reads lazygit pane for richer git context.
Stress and performance testing specialist. Finds what breaks under pressure through load testing strategies, performance profiling, edge case generation, and resource exhaustion scenarios. Use when you want to know what will break before it breaks in production. Triggers on "performance", "load test", "stress test", "what could break", "edge cases". SUGGEST PROACTIVELY WHEN: (1) new API endpoints or data processing functions written, (2) code handling concurrent requests, (3) database queries written (N+1 risk), (4) deploying to constrained envs (Pi 5, Workers 50ms CPU limit), (5) caching or rate limiting logic added.
A personal Claude Code marketplace with 3 plugins: 92 skills, 9 agents, and a full dev environment automation stack (hooks, coding standards, Claude Island integration).
git clone git@github.com:arosenkranz/claude-code-config.git ~/workspace/claude-code-config
~/.claude/settings.json{
"extraKnownMarketplaces": {
"arosenkranz-claude-plugins": {
"source": { "source": "github", "repo": "arosenkranz/claude-code-config" }
}
},
"enabledPlugins": {
"workflow-skills@arosenkranz-claude-plugins": true,
"goldeneye-agents@arosenkranz-claude-plugins": true,
"dev-environment@arosenkranz-claude-plugins": true
}
}
See config-templates/settings.json.template for a complete starting point.
cp config-templates/CLAUDE.md.template ~/.claude/CLAUDE.md
# Edit it — add your machine-specific paths, identity, infrastructure context
./scripts/migrate-to-marketplace.sh --dry-run # Preview
./scripts/migrate-to-marketplace.sh # Run
92 skills invocable via /skill-name:
/morning-plan /course-review /doc-generate
/api-scaffold /agent-browser /patrol
See plugins/workflow-skills/README.md for the full list.
9 specialized subagents (Operation Goldeneye roster):
| Agent | Role |
|---|---|
boris | Security specialist, attacker mindset |
m | Strategic planning, architecture, docs |
moneypenny | Session logging, blog content, brag docs |
natalya | TDD implementation engineer |
q | Infrastructure, Docker, CI/CD, incidents |
trevelyan | Adversarial code reviewer |
wade | Project continuity, session catch-up |
xenia | Performance and stress testing |
jira-manager | Jira ticket automation |
Lifecycle hooks, coding standards, and Claude Island integration:
session-start.sh, session-logger.sh, pre-compact.sh, post-edit-format.sh, check-console-log.sh, claude-island-state.py/dev-environment:coding-style, /dev-environment:git-workflow, /dev-environment:security, /dev-environment:testing, /dev-environment:performance, /dev-environment:memory-system| Tool | Purpose | Install |
|---|---|---|
| Claude Code | Required | See Anthropic docs |
| Node.js (v18+) | MCP servers via npx | brew install node |
| bats | Running tests | brew install bats-core |
| cmux | workspace skill | brew install cmux |
| yazi | workspace skill | brew install yazi |
| lazygit | workspace skill | brew install lazygit |
| agent-browser | browser automation | npm install -g agent-browser |
cmux, yazi, lazygit, and agent-browser are only required for specific skills.
These are never tracked by git — each machine maintains its own:
~/.claude/settings.json — model, permissions, plugins, MCP config~/.claude/CLAUDE.md — identity, machine paths, infrastructure contextStart from the templates in config-templates/.
bats tests/
Use this agent when creating user interfaces, designing components, building design systems, or improving visual aesthetics. This agent specializes in creating beautiful, functional interfaces that can be implemented quickly within 6-day sprints. Examples:\n\n<example>\nContext: Starting a new app or feature design
Uses power tools
Uses Bash, Write, or Edit tools
Comprehensive PR review agents specializing in comments, tests, error handling, type design, code quality, and code simplification
Upstash Context7 MCP server for up-to-date documentation lookup. Pull version-specific documentation and code examples directly from source repositories into your LLM context.
Use this agent when you need expert assistance with React Native development tasks including code analysis, component creation, debugging, performance optimization, or architectural decisions. Examples: <example>Context: User is working on a React Native app and needs help with a navigation issue. user: 'My stack navigator isn't working properly when I try to navigate between screens' assistant: 'Let me use the react-native-dev agent to analyze your navigation setup and provide a solution' <commentary>Since this is a React Native specific issue, use the react-native-dev agent to provide expert guidance on navigation problems.</commentary></example> <example>Context: User wants to create a new component that follows the existing app structure. user: 'I need to create a custom button component that matches our app's design system' assistant: 'I'll use the react-native-dev agent to create a button component that aligns with your existing codebase structure and design patterns' <commentary>The user needs React Native component development that should follow existing patterns, so use the react-native-dev agent.</commentary></example>
Comprehensive startup business analysis with market sizing (TAM/SAM/SOM), financial modeling, team planning, and strategic research
Semantic search for Claude Code conversations. Remember past discussions, decisions, and patterns.