generate-infra-code
Generate Kubernetes manifests, Terraform modules, Helm charts and the quality-gate CI workflow into generated/.
Skill by wfukatsu
Install
Works with Claude Code, Claude apps, Codex, Cursor, Gemini CLI, GitHub Copilot, VS Code, OpenCode, Goose, Windsurf. Read the repository before you give any tool access to your data.
npx agentmarket install skill/wfukatsu-generate-infra-code
Asks which of your tools to add it to (Claude Code, Codex, Cursor, VS Code, Windsurf, Gemini CLI, Claude Desktop) and shows the change first. See https://agentmart.dev/cli.
Get the weekly diff
The weekly diff: one email a week with new agent tools, notable updates and grade changes, plus security alerts for tools you follow. Double opt-in, one-click unsubscribe.
About
Use cases: devops, productivity
Reviews
Write a reviewNo reviews yet. Installed it? Reviews come from verified installers only. How reviews work
Permissions
What this skill can do and what it asks for, from its SKILL.md and code. Checked Oct 25, 2026.
- Tools
- No tools found
- Secrets
- None asked for
- Install scripts
- None found
Network hosts it talks to 1
codeload.github.com· found in install command
Files it touches
~/.claude/skills/generate-infra-code
What's inside
- Skill name
generate-infra-code- Description
- Generate Kubernetes manifests, Terraform modules, Helm charts and the quality-gate CI workflow into generated/. For merge-bound code in a real infrastructure repository, use /infra:implement.
- License
- MIT
- Instructions
- 91 lines in SKILL.md · view on GitHub
Agent Skills spec issues
- name: missing
Files (1)
- SKILL.md
SKILL.md (first lines)
# Infrastructure Code Generation
Shared files: a path written `@rules/…`, `@skills/…`, `@templates/…` or `@docs/…` is relative to the
plugin root, `${CLAUDE_PLUGIN_ROOT}` — not to the project being worked on. Read it from there.
## Desired Outcome
Generate IaC code based on the infrastructure design:
- Kubernetes manifests (Kustomize base + overlays)
- Terraform modules (multi-cloud support)
- Helm values (for ScalarDB Cluster)
- NetworkPolicy and PodDisruptionBudget
- Multi-environment configuration (dev/staging/prod)
- **The CI workflow that runs the quality gate** (see below)
## Quality Gate CI
The in-session gate that `implement-backlog` runs (@rules/ai-code-quality-gate.md) is fast feedback.
**The CI run is the one that is actually enforced** — it holds for hand-written changes, for changes
made months later, and for the times nobody thought to run it. A gate that exists only as a checklist
a model is asked to follow is the weakness this pipeline is closing, so emit the CI half whenever the
project has CI.
Generate a workflow for the project's platform (GitHub Actions / GitLab CI — detect it from the repo,
never assume), running the eight stages as jobs:
| Job | Runs |
|-----|------|
| build | The project's real build target |
| unit | The unit test task with coverage verification (`jacocoTestCoverageVerification`), and the mutation run (`pitest`) scoped to `domain/` — thresholds per @rules/ai-code-quality-gate.md §Test quality, read from the build files, never re-typed in the workflow |
| contract | The named contract-test task `generate-contract-tests` wired into the build |
| integration | The integration test task, including the transaction scenarios (OCC conflict, 2PC failure, saga compensation) |
| sast | Semgrep, or the project's configured scanner |
| dependency-scan | OSV-Scanner / Dependency-Check / `npm audit`, plus Gitleaks over the change |
| image-scan | Trivy, when the pipeline builds an image |
Rules the generated workflow must satisfy:
- **A stage that cannot run fails loudly.** Never emit `continue-on-error` or `|| true` on a gate job —
a green pipeline that skipped its scanners is worse than a red one, because it is read as evidence.
- **Never invent a command.** Every task named in the workflow is verified to exist in the project's
build files first; a missing target is reported as a gap, not papered over with a command that
happens to work.
- **Pin the actions and tool versions** per @rules/dependency-versions.md — a floating action tag is
an unpinned dependency in the security path.
- **Publish the evidence.** Each job uploads its report so the gate result is inspectable after the
fact, which is what makes stage evidence meaningful rather than ceremonial.
Stages 7–8 (API security, design↕code conformance) are model-driven and are **not** emitted as CI
jobs; the workflow records that they are covered in-session by
`/architect:verify-implementation --gate`, so their absence from CI is visible rather than silent.
## Dependency Versions
IaC is almost entirely version pins — Kubernetes `apiVersion` and cluster version, Terraform
`required_version` and provider constraints, Helm chart versions, container image tags. Follow
@rules/dependency-versions.md: resolve each from its registry (Terraform registry, chart repo, image
registry, `endoflife.date` for the supported Kubernetes window), pin **stable, non-EOL, mutually
compatible** versions, never leave a moving `:latest`/`stable` image tag in a manifest, and honour
what the target environment already runs over what is newest. Confirm the version decision table per
`--confirm-versions` / `--no-confirm-versions` / `options.confirm_versions`, and record it with the
generated IaC plus `work/version-decisions.json`.
## Prerequisites
| File | Required/Recommended | Source |
|------|---------------------|--------|
| reports/08_infrastructure/ | Required | /architect:design-infrastructure |
| reports/03_design/target-architecture.md | Recommended | /architect:design-microservices |
## Output
Write all reports in the language configured in `work/pipeline-progress.json` (`options.output_language`).
| File | Content |
|------|---------|
| `generated/infrastructure/k8s/` | Kubernetes manifests |
| `generated/infrastructure/terraform/` | Terraform modules |
| `generated/infrastructure/helm/` | Helm values |
| `generated/infrastructure/ci/` | The quality-gate workflow for the detected CI platform |
## Related Skills
| Skill | Relationship |
|-------|-------------|
| /architect:design-infrastructure | Input source |
| /architect:generate-contract-tests | Input source — the named contract-test task the CI workflow invokes |
| /architect:verify-implementation | Related — runs the same gate in-session; the CI workflow is its enforced half |
| /architect:generate-docs | Downstream — run after generation to write the README and the `operations` docs for the emitted IaC |From wfukatsu/nexus-architect, MIT.
More from wfukatsu/nexus-architect
- 0.632score
config
wfukatsu
Generate ScalarDB configuration files (database.properties, build.gradle dependencies) based on your choices.
SkillAFreshness: A★ 13Claude Code +9 - 0.632score
crud-ops
wfukatsu
Show ScalarDB CRUD operation patterns — builder syntax for Get, Scan, Insert, Upsert, Update, Delete with examples.
SkillAFreshness: A★ 13Claude Code +9 - 0.632score
docs
wfukatsu
Search ScalarDB / ScalarDL documentation for specific topics.
SkillAFreshness: A★ 13Claude Code +9 - 0.629score
analyze
wfukatsu
Extract ubiquitous language, actor-role-permission matrix, and domain-code mapping.
SkillAFreshness: A★ 13Claude Code +9 - 0.629score
build-app
wfukatsu
Build a complete ScalarDB application from a domain description.
SkillAFreshness: A★ 13Claude Code +9 - 0.629score
capture-followup
wfukatsu
Capture follow-up work discovered during backlog delivery — deferred tasks, out-of-scope findings, doc drift, waived acceptance criteria — into a reviewable…
SkillAFreshness: A★ 13Claude Code +9 - 0.629score
generate-test-specs
wfukatsu
Generate BDD scenarios, contract test, unit test, property-based test, integration test, and performance test specifications.
SkillAFreshness: A★ 13Claude Code +9 - 0.629score
analyze-data-model
wfukatsu
Comprehensive data layer analysis including entity/relationship analysis, DB design evaluation, and ER diagram generation.
SkillAFreshness: A★ 13Claude Code +9 - 0.629score
deliver-backlog
wfukatsu
Drive each Issue under an Epic through implement → review → human approval → merge, in order, resuming from backlog-manifest.json and stopping at every human…
SkillAFreshness: A★ 13Claude Code +9 - 0.629score
analyze-ui
wfukatsu
Analyze the user interface of an existing system: screens with their inputs, outputs, actions and access; UI components; features and user tasks; business…
SkillAFreshness: A★ 13Claude Code +9 - 0.629score
design-api
wfukatsu
Generate REST/GraphQL/gRPC/AsyncAPI specifications as the project's single, verifiable API contract — named schemas, every status code declared, RFC 9457…
SkillAFreshness: A★ 13Claude Code +9 - 0.629score
design-data-layer
wfukatsu
Generic database design for non-ScalarDB projects.
SkillAFreshness: A★ 13Claude Code +9
Compare and alternatives
Featured in
Signals
- Stars
- 13
- Forks
- 1
- Open issues
- 0
- Last commit (this folder)
- Oct 8, 2026 (19 days ago)
- Last release
- Oct 9, 2026 (18 days ago)
From GitHub (fetched Oct 24, 2026).
Details
- License
- MIT
- Repository
- github.com/wfukatsu/nexus-architect
- First seen
- Oct 24, 2026
- Last checked
- Oct 24, 2026
- Source
- GitHubgithub-graphql
More from wfukatsu
- 0.632score
docs
wfukatsu
Search ScalarDB / ScalarDL documentation for specific topics.
SkillAFreshness: A★ 13Claude Code +9 - 0.632score
crud-ops
wfukatsu
Show ScalarDB CRUD operation patterns — builder syntax for Get, Scan, Insert, Upsert, Update, Delete with examples.
SkillAFreshness: A★ 13Claude Code +9 - 0.632score
config
wfukatsu
Generate ScalarDB configuration files (database.properties, build.gradle dependencies) based on your choices.
SkillAFreshness: A★ 13Claude Code +9 - 0.629score
analyze
wfukatsu
Extract ubiquitous language, actor-role-permission matrix, and domain-code mapping.
SkillAFreshness: A★ 13Claude Code +9 - 0.629score
build-app
wfukatsu
Build a complete ScalarDB application from a domain description.
SkillAFreshness: A★ 13Claude Code +9 - 0.629score
analyze-ui
wfukatsu
Analyze the user interface of an existing system: screens with their inputs, outputs, actions and access; UI components; features and user tasks; business…
SkillAFreshness: A★ 13Claude Code +9
Cite this page
generate-infra-code skill. Agent Mart. https://agentmart.dev/skills/wfukatsu-generate-infra-code (last updated 2026-10-24)