Silver Platter — AI skill for Claude Code
The /silver-platter skill.
How to install Silver Platter
This entry records only its repository, not the path inside it, so there is no
exact command to give. Open earlyaidopters/silver-platter and copy the folder into
~/.claude/skills/, or the file into ~/.claude/agents/.
What Silver Platter does
The /silver-platter skill. Maps your back of house before you layer AI on top. Pantry > Prep > Plate > Then Agents.
Alternatives in AI
- AI Readiness — AI 就绪度评估 — 单票 AI 暴露/卡位评级 + 三道 gate + Top 杠杆点 6.6k ★
- Cartographer — Claude Code plugin that maps and documents codebases of any size using parallel AI subagents 518 ★
- Heard — Jarvis for your coding agents — the voice layer for Claude Code, Codex, OpenClaw, Hermes & any AI workflow 189 ★
README
/silver-platter
A Claude Code skill that interviews any business owner about their day-to-day tools, maps their back of house into a visual data architecture, and generates a plain-English 30-day build plan for layering AI on top.
The whole thing rests on one idea: **80% of building an Agentic OS is data prep. The 20% AI layer is easy.** This skill is the 80%.

What it does
You run `/silver-platter` inside any Claude Code project. The skill:
**Silently audits** your project folder for existing Claude Code setup (`.claude/`, `silver_platters/`, `data/`, hooks, agents, rules). If you've already started building, it skips questions about what's there.
**Interviews you** in plain English about your tools and workflow. No jargon. It asks what you use, how big your audience/customer base is, and what weekly task you'd most want to automate. It never asks you about "format", "cadence", "volume", or anything that sounds like a database column.
**Builds a data map** using the Pantry / Prep / Plate metaphor:
- Pantry = every place your raw data already lives (Shopify, your inbox, your EHR, your POS)
- Prep Table = weekly summary files ("silver platters") that Claude will write for you
- Plate = the briefs and drafts that land in front of you Monday morning
**Generates recipes** showing how your tools combine for specific business outcomes. Each recipe has: a headline you can screenshot, a "today, manually" movie scene of how you do it by hand, a "next Monday with this live" description, and a time-saved estimate.
**Generates a 30-day build plan** with 5 prioritized steps (conversion hook, silver platters, orchestrator + specialists, audit log + approval gates, slash commands). Each step has "you'll know it's working when" so you can verify yourself.
**Maps your interaction l
Related Skills
Prep Multipov
Verify that the multipov.ai MCP server is configured, authenticated, and responsive in Claude Code. Run before
Codeblast
Know what breaks before you merge — mutation-tested code graph with architecture, change & impact maps. Eviden
Tma1
Local-first observability your agent reads back. TMA1 records every LLM call, then routes what it sees into th
AI Citation Monitor
Tracks whether your brand is cited by ChatGPT, Claude, Perplexity, and Gemini for your target queries. Generat
Crackerjack Run
Run crackerjack with AI-powered auto-fixing using intelligent session integration. Automatically tries enhance
Refresh Coder Comparison
Re-research the coder capability matrix — re-check benchmark boards, pricing, model rosters, safety evaluation
Related Agents
Credential Hunter
Autonomous credential-attack pipeline runner. Chains /wordlist-gen + /osint-employees + /breach-check (data-pr
Business Process Architect
L0.7 Business-process agent. Dispatched under the CPO before a product/feature wave. Maps the CLIENT's current
Langertha Reviewer
Read-only code reviewer for Langertha — reviews a diff, branch or fix wave against its spec, plan, ADRs and ho