Claude Code · Enterprise PoC Stack

j-stack

A Claude Code stack optimized for enterprise PoC delivery.

01

Higher-fidelity demos in less time.

02

Defensive deliverables that preempt "did you try X" objections.

03

Cross-tool memory — resume in Codex / Cursor / ChatGPT when Anthropic limits hit.

The building blocks
Claude Code foundation

Anthropic's AI coding CLI. Not autocomplete — an autonomous agent that reasons about architecture, writes and edits files, runs tests, browses docs, and dispatches parallel subagents. Think of it as a senior engineer who lives in your terminal.

Superpowers framework

A plugin framework that installs structured workflow skills into Claude Code. Skills enforce how Claude approaches problems: brainstorm before planning, spec before building, test before shipping. j-stack uses brainstorming, writing-plans, TDD, and subagent-driven-development.

gstack skill pack

A community skill pack with 40+ specialized tools: security audits, CEO-lens scoping, design QA, document generation. j-stack cherry-picks 11 skills that complement Superpowers and injects explicit model directives so expensive models aren't wasted on mechanical work.

j-stack is the opinionated assembly. Raw Superpowers + gstack is powerful but sprawling — 40+ skills with no lane ownership, no model discipline, and no cross-tool memory. j-stack curates the set, routes models by cognitive demand, adds the four custom skills that complete the pipeline, and wires the whole thing into a repeatable workflow.

What it is

An opinionated assembly of three tools — curated, model-routed, and extended with four custom skills that don't exist anywhere else.

j-stack assembles Claude Code + Superpowers + cherry-picked gstack skills + four custom skills into an opinionated workflow pipeline. Each phase is owned by a specific skill — nothing overlaps, nothing conflicts. The result is a repeatable system for shipping polished, defensible PoC artifacts.

Spec before build

Writing-plans locks a full implementation spec before any code runs. Subagents implement against the spec, not a vague prompt.

TDD during build

The TDD skill enforces red-green-refactor discipline. Tests are written first, against the spec, then implementation follows.

Cross-vendor defense

OpenAI Codex independently reviews Claude's work. Convergent findings from two AI vendors are the strongest signal you can put in front of a skeptical architect.

Context that survives limits

A .planning/ wiki keeps every decision readable by Claude Code, Codex, Cursor, ChatGPT, and Gemini. Hitting usage limits mid-engagement is a handoff, not a loss.

SDLC coverage — discovery to demo
SDLC Phase
Stage
Skills
PM Principle
Product Discovery EXPAND
/office-hours /plan-ceo-review
Are we solving the right problem? Founder/10x reframe before any engineering begins.
Requirements & Ideation REFINE
brainstorm (SP)
Structured pressure-testing before committing to an approach.
Feasibility / Build-vs-Buy SURVEY
prior-art-survey
Named, reasoned alternatives across OSS, libraries, and patterns. Every "did you try X" answered before the meeting.
Technical Design PLAN
writing-plans (SP)
Spec locked before implementation begins. Explicit gate — nothing builds until this is done.
Development BUILD
subagent-driven-dev TDD (SP) /design-shotgun
TDD enforced (red-green-refactor). Parallel subagents implement against the locked spec.
QA & Security Review POLISH
/qa /design-review /cso
Definition of done against spec. OWASP/STRIDE risk management as a deliverable, not a checkbox.
Stakeholder Review DEFEND
second-opinion stakeholder-pack
Pre-answered objections. Cross-vendor AI validation. Meeting shifts from "defend" to "what's next."
Documentation & Handoff HANDOFF
/document-release handoff-snapshot
Docs generated from diff, not written afterward. Knowledge transfer built in.
Out of scope by design
Deployment / Release ops Monitoring / Observability Incident response / Maintenance PoC mission ≠ production
Engineering PM principles baked in
Discovery before engineering

EXPAND + REFINE exist before a line of spec is written. The pipeline cannot reach PLAN without completing both.

Build vs. buy at every dependency

prior-art-survey dispatches three parallel scouts — OSS, libraries, patterns — and produces a named, reasoned landscape. Not a gut call.

Explicit phase gates

Each stage must complete before the next begins. Nothing in BUILD runs without a locked PLAN output.

Risk as a deliverable

/cso files an OWASP/STRIDE finding set. second-opinion adds a cross-vendor risk matrix. Both are artifacts, not chat notes.

Docs built in, not bolted on

/document-release generates docs from the diff. handoff-snapshot captures state. No documentation sprint at the end.

Parallel workstreams

subagent-driven-development and prior-art-survey's scouts both parallelize work the way a PM schedules concurrent engineering tracks.

Spec-driven · TDD by default
Plan before build

Nothing in BUILD starts until PLAN is locked. The pipeline enforces four phases before a line of implementation code is written: reframe the problem → pressure-test the approach → survey the prior art → spec the implementation.

writing-plans produces a full implementation plan reviewed by Opus. Subagents in BUILD implement against that spec — not a vague prompt, not a wish.

TDD during build

The superpowers:test-driven-development skill enforces red-green-refactor inside the build phase. Tests are written first against the spec; implementation follows only once the test is failing for the right reason.

subagent-driven-development parallelizes implementation across multiple Claude instances, each operating against the same locked spec and test suite.

Defensive by design — the five questions enterprise stakeholders always ask
"Why did you build it this way?"
/office-hours + brainstorm

Vision doc traces the founder-lens reframe and every major decision made during scoping.

"What else did you consider?"
prior-art-survey

Parallel OSS, library, and pattern scouts produce a named, reasoned landscape. "Did you try X" is answered before it's asked.

"Is it secure?"
/cso

OWASP Top 10 + STRIDE security review runs before the demo. Findings are filed by severity with recommended mitigations.

"Does it actually work?"
/qa + TDD suite

QA audits the running app against the spec. The TDD test suite from BUILD provides the passing-green baseline.

"Why should I trust one AI vendor?"
second-opinion

OpenAI Codex independently reviews Claude's work. A convergence/divergence matrix is the deliverable — two AI vendors agreeing is a stronger signal than one.

stakeholder-pack

All five answers assembled into one executive-ready document before the demo. The meeting shifts from "defend the work" to "discuss what's next."

Pipeline
j-stack pipeline · 8 stages
EXPAND
/office-hours /plan-ceo-review
REFINE
brainstorm (SP)
SURVEY
prior-art-survey
PLAN
writing-plans (SP)
BUILD
subagent-driven-dev /design-shotgun /design-html
POLISH
/qa /design-review /cso
DEFEND
second-opinion stakeholder-pack
HANDOFF
/document-release handoff-snapshot
hover a stage to inspect
Discovery
Planning
Execution
Delivery

SP = Superpowers

Stack layers
Superpowers core
brainstorming writing-plans subagent-driven-development
gstack 11 skills
/office-hours /plan-ceo-review /qa /design-shotgun /design-html /design-review /cso /codex /document-release /freeze /guard
Custom skills 4 skills
poc-wiki-init handoff-snapshot second-opinion stakeholder-pack
prior-art-survey scout
Parallel OSS scouts library pattern matching competitive landscape
Model routing
opus
claude-opus
Judgment moments
Scoping & discovery
Security review
Cross-model synthesis
Stakeholder framing
hku
claude-haiku
Housekeeping
Templating
Summarizing
Mechanical operations
Cross-tool memory

Never lose context at usage limits.

Each PoC project gets a .planning/ wiki with schema files for every major tool — CLAUDE.md, AGENTS.md, .cursor/rules, chatgpt-brief.md. When Anthropic limits hit, run handoff-snapshot and paste the continuation prompt into your next tool.

Context flow
Claude Code primary agent
sync
.planning/wiki shared memory
handoff
Codex
Cursor
ChatGPT
Gemini
schema files
CLAUDE.md AGENTS.md .cursor/rules chatgpt-brief.md
Install
bash
> cd j-stack
> bash install.sh