Replica Architect
Turns an app recon map into a concrete stack choice, SQL schema, API route table and a vertical-slice-first build order.
What it does
Reads replica/recon.md and replica/features.csv (produced by the companion /replica-recon skill) and writes replica/architecture.md plus replica/schema.sql.
- Stack selection: an opinionated table for web, mobile, database, ORM, auth, payments, email, jobs, file storage and hosting, each with a one-line rationale. One database, no microservices.
- Schema design: uuid primary keys,
created_at/updated_at, owner columns, expliciton deleterules, indexes on every FK and filter/sort column, check constraints for status enums,timestamptzin UTC, money as integer cents + currency, and either Postgres RLS or an explicit per-query authorisation check. - API design: one table per flow —
method path | what it does | who can call it | input | output— plus inbound/outbound webhooks and scheduled background jobs. - Sharp edges checklist: time zones and DST, webhook idempotency, race conditions, rate limits, search, realtime, offline, deliverability, multi-tenancy, GDPR deletion.
- Build order: a thin end-to-end vertical slice first, then must-haves, should-haves and could-haves, with screens, tables and routes per milestone.
Who it's for
- Indie hackers and founders rebuilding their own version of a familiar SaaS
- Junior-to-mid fullstack devs who stall at table design and endpoint inventories
- Small teams that want the shortest credible path to an MVP
Examples
- "Design the database for my Calendly clone" → a
bookingstable with timestamptz columns and a DB-level constraint that prevents double-booking. - "What stack should I use?" → a defaults table (Next.js, Supabase, Stripe, Vercel) with swap options and one-line reasons.
- Right after
/replica-reconfinishes → feature list reordered into milestones with the three riskiest parts called out.
Note: the skill explicitly forbids calling the original app's private endpoints and requires official public APIs with your own keys.
· · · Install guide · · ·
Try it now, no install
Paste this into Claude to use the skill without installing anything.
Read the instructions in this file and follow them to help me: https://raw.githubusercontent.com/Jakeschincariol/replica-skill/HEAD/replica-architect/SKILL.md What I want: (describe your task here)
If Claude can't open the link, open it yourself and paste the contents instead.
↓ If it works for you, download the ZIP below and install it. Then it runs on its own — no pasting each time.
Install in the Claude app (no terminal)
- Download the ZIP with the button below.
- In Claude, open Settings → Capabilities and turn on 'Code execution and file creation'. (one time)
- Go to Customize → Skills → + → 'Upload a skill' and upload the ZIP.
Install in Claude Code
Let Claude do it — paste this into Claude Code
Install the skill I found on Claude Skill Mart. Copy the replica-architect folder from the GitHub repo Jakeschincariol/replica-skill into my ~/.claude/skills/replica-architect/. When it's done, tell me in one line what this skill can do.
Install with a command instead
git clone https://github.com/Jakeschincariol/replica-skill.git /tmp/replica-skill && mkdir -p ~/.claude/skills && cp -r /tmp/replica-skill/replica-architect ~/.claude/skills/⚠ This is a third-party skill. Check the source repository before installing.
- Open your terminal.
- Clone the repo:
git clone https://github.com/Jakeschincariol/replica-skill.git - Create the skills folder:
mkdir -p ~/.claude/skills - Copy the skill in:
cp -r replica-skill/replica-architect ~/.claude/skills/ - (Recommended) Also copy the companion:
cp -r replica-skill/replica-recon ~/.claude/skills/ - Restart Claude Code, then ask "plan the clone" / "design the database" or invoke
/replica-architect. - Run
/replica-reconfirst — withoutreplica/recon.mdthe skill stops by design.