Specialized agents for advanced Alliance development workflows. Includes MCP server architecture and PRD writing.
Use this agent when you need to design, implement, or troubleshoot Model Context Protocol (MCP) servers that bridge LLMs with external APIs for natural language querying. This includes creating new MCP integrations, configuring existing servers, debugging connection issues, or architecting search capabilities over API data. Examples: **Example 1 - Creating a new MCP server:** user: "I want to connect Claude to our internal inventory database so I can ask questions like 'what products are low in stock?'" assistant: "I'll use the mcp-architect agent to design and implement an MCP server for your inventory database." <Task tool call to mcp-architect agent> **Example 2 - Integrating a third-party API:** user: "Can you help me set up an MCP server that connects to the GitHub API so I can search issues and PRs with natural language?" assistant: "Let me launch the mcp-architect agent to create a GitHub MCP integration with natural language search capabilities." <Task tool call to mcp-architect agent> **Example 3 - Debugging MCP configuration:** user: "My PubMed MCP server isn't returning results when I ask about recent papers. Here's my config..." assistant: "I'll use the mcp-architect agent to diagnose and fix your MCP server configuration." <Task tool call to mcp-architect agent> **Example 4 - Proactive suggestion after API discussion:** user: "We have a REST API for our customer support tickets at api.company.com/tickets" assistant: "I notice you have a REST API that could benefit from natural language querying. Let me use the mcp-architect agent to design an MCP server that would let you ask questions like 'show me urgent tickets from this week' or 'find tickets about billing issues'." <Task tool call to mcp-architect agent>
Use this agent when you need to create a Product Requirements Document (PRD) for any feature or platform initiative. This agent should be used proactively whenever there's a high-level product request that needs to be transformed into a comprehensive, executive-ready PRD. The agent will produce a crisp, decision-friendly document with clear requirements, metrics, and implementation details. Examples: <example>Context: The user wants to develop a new feature for their application.user: "We need to add a recommendation engine to our e-commerce platform"assistant: "I'll use the product-manager agent to create a comprehensive PRD for this recommendation engine feature"<commentary>Since this is a high-level feature request that needs to be turned into a structured product document, use the product-manager agent to create a PRD.</commentary></example><example>Context: The user is planning a platform initiative.user: "Let's build a new API gateway for our microservices"assistant: "I'm going to use the product-manager agent to develop a detailed PRD for the API gateway initiative"<commentary>This is a platform initiative that requires proper product documentation, so the product-manager agent should be used to create the PRD.</commentary></example>
Uses power tools
Uses Bash, Write, or Edit tools
Own this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge. GitHub access is read-only (username + org membership).
Sign in to claimOwn this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge. GitHub access is read-only (username + org membership).
Sign in to claimBased on adoption, maintenance, documentation, and repository signals. Not a security audit or endorsement.
Shared Claude Code configurations, plugins, and best practices for Alliance of Genome Resources developers.
[!IMPORTANT] Claude Code runs in a regular terminal, not in the Claude web UI. Alliance access is for the Claude Code command-line tool only. If the Claude website or Claude desktop chat app still shows a personal
Freeplan, that is expected. To use Alliance-provided Claude Code access, open a normal terminal window on your computer (for example Terminal, iTerm, Windows Terminal, or PowerShell) and run Claude Code there.
Claude Code is a command-line tool that brings Claude AI directly into your terminal. Instead of copying code back and forth from a chat window, Claude Code can:
You interact with it through natural conversation in your terminal:
You: Can you add error handling to the API client?
Claude: I'll add try/catch blocks and proper error messages. Let me read the current
implementation first...
[Claude reads files, makes edits, runs tests]
To get started with Claude Code at the Alliance, you need to be added to our Anthropic organization:
You'll receive an email invitation to join the Alliance Anthropic organization once your request is processed.
⚠️ Please take 10 minutes to read this: Claude Code Quickstart Guide
The official guide covers essential workflows that will save you hours:
You'll get far more out of Claude Code if you understand its capabilities upfront.
As mentioned in the quickstart guide, once you've received your invitation and created your account, install the Claude Code CLI:
macOS, Linux, WSL:
curl -fsSL https://claude.ai/install.sh | bash
Windows PowerShell:
irm https://claude.ai/install.ps1 | iex
Alternative: Homebrew (macOS/Linux):
brew install --cask claude-code
One-time installation: Claude Code only needs to be installed once per computer. After installation, you can run
claudefrom any directory - you don't need to install it separately for each project or repository.
Note: Native installations auto-update automatically. Homebrew requires manual updates via
brew upgrade claude-code.
claude
> what does this project do?
⚠️ Important: Make sure your model is set to Opus for the best results.
Check and set your model:
/model
Select Opus from the list. This setting persists across sessions.
Why Opus? Opus is Claude's most capable model - it handles complex codebases better, makes fewer mistakes, and produces higher quality code. If you're finding Claude's suggestions unhelpful or off-target, check that you're using Opus.
These three habits will dramatically improve your experience with Claude Code. They're simple but make a huge difference in the quality and reliability of the code Claude produces.
npx claudepluginhub alliance-genome/agr_claude_code --plugin alliance-agentsSecret scanning pre-commit hooks using Gitleaks and TruffleHog. Blocks commits containing API keys, passwords, and tokens.
Comprehensive Jira management for Alliance projects (KANBAN, SCRUM, AGRHELP, MOD). Create tickets, search issues, manage status transitions, and more.
Expert LinkML schema reviewer for Alliance of Genome Resources. Reviews schema changes for correctness, consistency, and adherence to Alliance conventions.
Transform Claude Code into a structured development platform with 29 /sc: commands, 23 specialized agents, 7 behavioral modes, and MCP server integration
Multi-agent /workflow development pipeline (planner → plan-review → coder → code-review) with typed handoff contracts, lifecycle hooks, and MCP servers.
High-intelligence Claude Code copilot with deep code reasoning, evidence-driven planning, orchestration-first execution, model routing, context budgeting, CI/CD integration, enterprise security, plugin development, prompt engineering, performance profiling, agent teams, channels (event-driven autonomy with CI webhook, mobile approval relay, Discord bridge, and fakechat dev profile), interactive tutorials, LSP integration, security-hardened hook script library, MCP Prompts coverage, common workflow packs, runtime selection guide, computer-use patterns, checkpointing, scheduled-task blueprints, repo bootstrap scanner, hook policy engine (8 installable packs), layered memory deployment, role-based subagent packs (implementer, debugger, migration-lead, dependency-auditor, release-coordinator), 5 agent-team topology kits, autonomy operating mode (4 profiles + 3 gates), and a queryable 15-tool MCP documentation server with autonomy advisor.
Claude Code workflow patterns: prompting, CLAUDE.md maintenance, multi-agent orchestration
Production-ready Claude Code configuration with role-based workflows (PM→Lead→Designer→Dev→QA), safety hooks, 44 commands, 19 skills, 8 agents, 43 rules, 30 hook scripts across 19 events, auto-learning pipeline, hook profiles, and multi-language coding standards
MCP server development helper with tool and resource scaffolding