Claude Full Stack 2.0
AI-augmented software engineering skills from idea to production.
A Claude Code plugin: a curated set of production-grade Claude Skills that take a software project from idea to production. Positioned as AI-augmented software engineering — not a prompt collection — with primary differentiation in DevOps, SRE, and production operations.
Install
Add the marketplace, then install the plugin from it:
/plugin marketplace add amritmalla/claude-full-stack-2.0
/plugin install claude-full-stack-2-0@amritmalla-plugins
Once installed, the skills below become invocable by Claude Code's Skill tool whenever their description matches your request.
Repository layout
This repo is an AI-augmented engineering operating system. Skills are organized by what they own, not by topic:
- skills/architecture/ — technology-agnostic decision domains (product, architecture, backend, security, etc.).
- skills/implementations/ — ecosystem-specific execution (Spring Boot, Postgres, Kubernetes, GitHub Actions, ...).
- architecture-patterns/ — reusable architectural strategies (event-driven, hexagonal, modular-monolith, ...).
- standards/ — shared interoperability contracts that everything above conforms to.
- workflows/ — end-to-end execution flows that chain architecture + implementations.
See the long-form rationale in docs/architecture/research.md and the guiding beliefs in docs/philosophy/.
Skills
The repository ships 84 skills and 4 workflows across the architecture decision domains and ecosystem implementations (Spring Boot, FastAPI, Node.js, React, Flutter, Postgres, MongoDB, Kubernetes, AWS, Terraform, GitHub Actions, OpenAI, Anthropic, LangChain, AutoGen, CrewAI, ...). Browse the full catalog under skills/architecture/ and skills/implementations/.
A representative end-to-end lifecycle path on the Spring Boot reference stack:
| Stage | Skill | What it produces |
|---|---|---|
| Validation | saas-idea-validation |
Evidence-scored validation brief with a proceed / pivot / kill verdict (SaaS ideas; optional) |
| Idea | idea-development |
Discovery, refinement, validation, PRD specification, and execution readiness |
| Architecture | system-design |
System design + ADRs |
| Backend architecture | backend-architecture |
Backend boundaries, domain behavior, contracts, transactions, handoff notes |
| Quality | quality-engineering |
Contract-driven test strategy, acceptance criteria, and integration test plan |
| Operations | operations |
Blameless postmortem, reusable runbook, and on-call handoff |
| Backend scaffold | spring-boot-service-scaffold |
Production-ready Spring Boot layout |
| Security | spring-security-auth-review |
JWT/OAuth2 review and hardening |
| Observability | spring-boot-observability-readiness |
SLIs/SLOs + multi-burn-rate alerts |
| Data | postgres-schema-and-migration |
Schema + zero-downtime migration plan |
| Containers | dockerfile-and-jvm-tuning |
Multi-stage Dockerfile + container-aware JVM flags |
| CI/CD | github-actions-pipeline-hardened |
OIDC, pinned SHAs, SBOM, cosign |
| Deploy | k8s-deploy-manifest-review |
Hardened Kubernetes manifests |
Workflows
Workflows sequence skills into lifecycle paths with explicit Entry/Exit/Gate checkpoints per phase:
idea-to-production-full-stack— Spring Boot + Postgres backend and React web frontend, idea to production on Kubernetes (the backend+web capstone).idea-to-production-flutter— a Flutter mobile app from idea to signed store release.production-readiness-review— a non-build hardening pass (reliability → security → performance → operations) for an existing service.cloud-foundation-on-aws— an infrastructure-only AWS landing zone and Terraform delivery foundation.
Claude Desktop
Claude Desktop doesn't load Claude Code plugins, but you can expose the 83 skills + 4 workflows as MCP tools via the companion package:
pip install claude-full-stack-skills
Then add to your claude_desktop_config.json:
{
"mcpServers": {
"claude-full-stack-skills": {
"command": "claude-full-stack-skills"
}
}
}
Restart Claude Desktop. Skills become invocable by description match — same
routing logic as Claude Code's Skill tool. Full setup steps and reference-file
lookup are documented in INTEGRATION.md.
Memory
For durable cross-session memory and hierarchical retrieval over a repo's code, docs, and decisions, install claude-repo-mem and follow INTEGRATION.md. The memory-management skill defines when each claude-repo-mem tool (recall, trace, remember, plan_task, handoff, resume, ...) is appropriate during architecture, implementation, and operations work.
Scope
In: the full "Full Stack 2.0" surface — architecture decision domains plus ecosystem execution across backend (Spring Boot, FastAPI, Node.js), frontend (React), mobile (Flutter), data (Postgres, MongoDB), infrastructure (Kubernetes, AWS, Terraform, GitHub Actions), and AI (OpenAI, Anthropic, LangChain, AutoGen, CrewAI), composed by end-to-end workflows.
Not yet: several ecosystem directories are taxonomy placeholders awaiting skills (e.g. Go, Django, .NET, Vue, Azure, GCP).
Companion package: claude-full-stack-skills (source under src/claude_full_stack_skills/) is the MCP server that exposes every skill + workflow as an MCP tool for Claude Desktop and other MCP hosts. See the Claude Desktop section above.
Contributing
Read CONTRIBUTING.md and CONVENTIONS.md. New skills follow SKILL_SPEC.md. New workflows follow WORKFLOW_SPEC.md.
License
MIT. See LICENSE.
Metadata
Release files for claude-full-stack-skills 0.3.0
For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.
Source distribution (sdist)
| File | Size | Uploaded | |
|---|---|---|---|
| claude_full_stack_skills-0.3.0.tar.gz | 600.0 kB | Details |
Built distribution (wheel)
| File | Interpreter | ABI | Platform | Reset |
|---|---|---|---|---|
| claude_full_stack_skills-0.3.0-py3-none-any.whl | Python 3 | none | any | Details |
Total release size: 1.6 MB
Release files / claude_full_stack_skills-0.3.0.tar.gz
| Download URL | claude_full_stack_skills-0.3.0.tar.gz |
|---|---|
| Size | 600.0 kB |
| Tags | Source |
|
SHA-256 checksum How to use checksums |
61471da3c3b9aee4471f110ff6f5bbc7453f611adbc0a039e5542f290fa050ae
|
|
BLAKE2b-256 checksum How to use checksums |
cc63570bf5ecb0674b7ae21c4f61f95a6e917f26f5ccfea282b659542045d7f5
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
Yes |
| Uploaded via |
twine/7.0.0 CPython/3.13.14
|
Provenance
Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.
PyPI Publish Attestation
PyPI verified that this artifact, at this checksum, originated from the publisher listed below.
Signed by GitHub Actions, verified by PyPI on Aug 8, 2026.
Transparency logRelease files / claude_full_stack_skills-0.3.0-py3-none-any.whl
| Download URL | claude_full_stack_skills-0.3.0-py3-none-any.whl |
|---|---|
| Size | 954.0 kB |
| Tags | Python 3 |
|
SHA-256 checksum How to use checksums |
c1078a5a061e2f727098d819643cb271b195a046874328e56e8b6542e0f10b8c
|
|
BLAKE2b-256 checksum How to use checksums |
6163958d7f62d9f36312a17b1021e1af74396bce4caecca54c60f9bf889bcdd5
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
Yes |
| Uploaded via |
twine/7.0.0 CPython/3.13.14
|
Provenance
Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.
PyPI Publish Attestation
PyPI verified that this artifact, at this checksum, originated from the publisher listed below.
Signed by GitHub Actions, verified by PyPI on Aug 8, 2026.
Transparency log