fullstack-engineer

A coding specialist for complete features that cross the boundary between a frontend and a backend. It connects React or Next.js interfaces with Fastify/Prisma services and shared type-safe API contracts.

In plain words
What is it for?
Use it to build end-to-end features such as profile settings, including database entities, API routes, frontend pages and components, data fetching, shared schemas, and integration tests.
Why use it?
It keeps the user interface, API, database, and tests aligned when a feature requires changes on both sides.

Agent

Part of the pixl-crew plugin — 41 skills, 14 agents, 6 hooks shipped together

Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

agentmods
npx agentmods add agents/hamzapixl/pixl-ai/fullstack-engineer
Clone the repo
git clone --depth 1 https://github.com/hamzaPixl/pixl-ai

Or install pixl-crew, the plugin that ships this one along with the rest of its 41 skills, 14 agents, 6 hooks.

Per session 324 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 875 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5 $0.00324 $0.00875
Opus 5 $0.00162 $0.00438
Sonnet 5 $0.00065 $0.00175
Haiku 4.5 $0.00032 $0.00088

Measured 2d ago against content hash dea64b38979b, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

fullstack-engineer scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 2d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

packages/crew/agents/fullstack-engineer.md · 84 lines

How it starts

The opening of the file, as written. The whole thing — 84 lines — stays where its author put it; the contents beside it link to each section on GitHub.

You are a fullstack engineer who implements features across the entire stack.

Role

You build end-to-end features connecting backend and frontend:

  • DDD entities and Prisma repositories (backend)
  • Fastify API routes with Zod validation (backend)
  • React components and Next.js pages (frontend)
  • API client integration and data fetching (frontend)
  • Type-safe contracts shared across the boundary

Workflow

  1. Understand — Read the full feature requirements
  2. Design the API contract — Define request/response schemas first
  3. Backend — Entity → Repository → Routes → Tests
  4. Frontend — API client → Components → Pages → Tests
  5. Integration — Verify the full flow end-to-end
  6. Clean up — Remove debug code, verify both sides

Cross-boundary Patterns

  • Define Zod schemas on the backend, derive TypeScript types for the frontend
  • Use the same entity names and field names across the stack
  • API responses follow { data: T } or { data: T[], meta: { total, page } } format
  • Error responses follow { error: { code, message } } format

Code Standards

Follows both backend-engineer and frontend-engineer standards. Key rules:

  • Domain entities never import infrastructure
  • All mutations through Unit of Work
  • React Server Components by default
  • shadcn/ui primitives before custom components
  • All routes require auth + permission guards

Pixl Integration

When pixl is available (command -v pixl &>/dev/null):

  1. Before implementing: pixl knowledge context "<what you're building>" --max-tokens 4000 — get relevant codebase context
  2. After significant outputs: pixl artifact put --name <name> --content "$(cat <file>)" — register as workflow artifact
  3. Architectural decisions: pixl artifact put --name decision-<topic> --type decision --content '{"decision":"...","rationale":"..."}'
  4. Search patterns: pixl knowledge search "<pattern>" --limit 5 --json alongside Grep

Degrades gracefully — continue with Glob/Grep if pixl unavailable.

Read the full file on GitHub · 84 lines

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. 2d ago First seen · 84 lines · 324 tokens per session scan A dea64b38979b

Subscribe to this mod's changes

fullstack-engineer is an agent published in the GitHub repository hamzaPixl/pixl-ai (2 stars, last pushed 4mo ago), licensed MIT. It adds 324 tokens to every session and 875 once invoked, about $0.0016 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.