ai-agent-project-scaffoldlisted

Stand up a new AI/agent project - or add an AI subsystem to an existing repo - starting from a requirements intake and ending in a committed full stack. Use when the user says "scaffold an agent project", "bootstrap a RAG/agent repo", "add an agent to this codebase", "add an AI/LLM feature to our existing app"; when they ask for the stack for a specific system they intend to build ("get me a stack", "which stack should we use"), as opposed to open-ended options questions, which go to ai-engineering; when they make ANY "build me an agent that does X" request, because the intake may rightly conclude a plain script suffices; and when a vague AI build idea needs shaping first, in which case it calls interview-me and idea-refine. Hands implementation to agent-skills.
thefilesareinthecomputer/dotagents · ★ 0 · AI & Automation · score 72

Install: claude install-skill thefilesareinthecomputer/dotagents

# AI Project Scaffold Turns an AI build request into a **grounded, hardened project skeleton** - but only after the requirements are clear enough to choose a stack. The orchestrator (you) runs the intake; the `ai-engineering` map picks the components; the `ai-engineer` subagent does the heavy build; `agent-skills` enforce build discipline. ## Step 1 - Intake (do not skip) This is the AI-specific extension of general elicitation. Call `interview-me` when the idea is still vague enough that the fields below cannot be answered, and `idea-refine` to stress-test a premise that sounds shaky. Those skills get the idea into focus; this one carries it into a committed stack, which they do not cover. Gather these before proposing any stack. Use `AskUserQuestion` for choices. Fill what's inferable from the repo or context; ask only for the rest. Never fabricate a constraint. 1. **Outcome** - what must the AI system *do*? (the job, in plain terms, and the definition of done) 2. **Surface** - CLI · backend service/API · in-app copilot/UI · batch/pipeline · MCP server. 3. **Model strategy** - local-first (Ollama / OpenAI-compatible) · cloud (Anthropic/OpenAI/Gemini) · hybrid with escalation. Data sensitivity / residency constraints. 4. **Footprint & lock-in** - existing cloud/stack to fit into? Lock-in tolerance? OSS-only mandate? 5. **Layers needed** - orchestration (always) · memory? · retrieval/RAG? · ingestion (web/docs)? · tools/MCP? · frontend? · eval/observability? 6. **Scale /