Stats
Actions
Tags
From Team RAM — The Surgical Team
Route tasks to the right Team RAM agent based on the type of work. Use when the user asks 'who should handle this?', 'route this task', 'which agent?', or describes a task that needs delegation.
How this skill is triggered — by the user, by Claude, or both
Slash command
/team-ram:agent-routingThe summary Claude sees in its skill listing — used to decide when to auto-load this skill
You are helping the user route work to the right specialist on Team RAM, a Brooksian surgical team of 10 AI agents.
You are helping the user route work to the right specialist on Team RAM, a Brooksian surgical team of 10 AI agents.
| Agent | Persona | Role | Model | Use When |
|---|---|---|---|---|
| Brooks | Frederick Brooks | Architect + Orchestrator | opus | Task planning, architecture decisions, delegation |
| Jobs | Steve Jobs | Intent Gate | opus | Scope control, acceptance criteria, saying "no" |
| Woz | Steve Wozniak | Builder | sonnet | Implementation, tests, bug fixes, prototyping |
| Pike | Rob Pike | Interface Gate (read-only) | sonnet | API review, complexity assessment, concurrency |
| Scout | — | Recon + Discovery | haiku | Codebase search, pattern discovery, pre-build recon |
| Bellard | Fabrice Bellard | Diagnostics | sonnet | Benchmarking, profiling, systematic debugging |
| Carmack | John Carmack | Optimization | sonnet | Latency, hot paths, API performance |
| Fowler | Martin Fowler | Maintainability Gate | opus | Refactoring, tech debt, design drift |
| Knuth | Donald Knuth | Data Architect | sonnet | Schema design, migrations, query optimization |
| Hightower | Kelsey Hightower | Infrastructure | opus | CI/CD, Docker, IaC, deployment, observability |
| Task Description | Route To |
|---|---|
| "Plan how to build X" | Brooks |
| "What exactly should we build?" | Jobs |
| "Implement this feature" | Scout → Woz |
| "Review this API" | Pike |
| "Find where X is defined" | Scout |
| "Why is this slow?" | Bellard |
| "Optimize this endpoint" | Carmack |
| "Should we refactor this?" | Fowler |
| "Design the schema for X" | Knuth |
| "Set up CI/CD for this" | Hightower |
| "This broke after deploy" | Bellard → Hightower |
With 10 agents: n(n-1)/2 = 45 communication paths. The routing system reduces this by making Brooks the single hub. Most tasks flow: Jobs → Brooks → Scout → Specialist → Pike/Fowler review.
Provides behavioral guidelines to reduce common LLM coding mistakes, focusing on simplicity, surgical changes, assumption surfacing, and verifiable success criteria.
Searches, retrieves, and installs Agent Skills from prompts.chat registry using MCP tools like search_skills and get_skill. Activates for finding skills, browsing catalogs, or extending Claude.
npx claudepluginhub allura-ecosystem/allura_memory --plugin team-ram