claude

nolte-engineering

Engineering capabilities for code-bearing projects: full-stack implementation, the test-tier and test-cycle suite, the quality gate, frontend/web-UI optimization, and code-security / dependency / license auditing. Code repositories adopt this on top of nolte-shared.

0 on GitHubby noltev0.1.11MITupdated 3 days ago
/

Plugin nolte-engineering from marketplace nolte-shared in nolte/claude-shared. Commands read from the repo's manifest — 2 days ago.

What it installs

Skills (19)

  • api-documentation-audit

    Audits a repository''s OpenAPI documentation against spec/project/api-documentation/ and produces a severity-classified audit artifact. Dispatches the read-only api-documentation-scanner agent to discover every OpenAPI document (checked in at a conventional location or exportable via a documented co

  • api-error-check

    Statically checks a web API''s error-handling surface for conformance against the project''s own declared error contract — uniform error-body shape, populated required fields, a dynamically-generated error/correlation id, correct HTTP status-code semantics, and no internal-detail leakage (stack trac

  • bdd-scenario-generate

    Generates executable BDD scenarios from an abstract test-case document (TC-IDs plus behaviors) per spec/project/behavior-driven-development/, emitting English Gherkin `.feature` files and thin `pytest-bdd` step-definition skeletons that honor the page-object decoupling contract of spec/project/bdd-p

  • capability-reach-audit

    Measures capability reach per spec/project/capability-reach-audit/ in a local working copy. `derive` drafts probes via the read-only capability-reach-scanner and persists only approved ones under project/reach-probes/; `run` executes them through bundled reach_audit.py (T0, T1; T2 on request), surfa

  • dependency-audit

    Scans the current project's dependency tree for known vulnerabilities (CVEs) and, when requested, license-compliance issues. Dispatches dependency-audit-scanner agent for the read-only scan step. Detects project kind from `pyproject.toml` / `requirements*.txt` / `poetry.lock` / `uv.lock` for Python

  • deployment-chart-manage

    Provisions and maintains a Helm application chart on the bjw-s `common` library, applying spec/project/bjw-s-common-chart-deployment/ and spec/project/kubernetes-deployment-best-practices/. Two operations: `provision` generates a complete best-practice chart greenfield; `reconcile` is the change-not

  • dockerfile-audit

    Audits a project's Dockerfiles against spec/project/dockerfile-best-practices/ and produces a severity-classified audit artifact. The default `audit` operation dispatches the read-only dockerfile-audit-scanner agent, then hard-fails on any missing mandatory OCI core label (source, title, description

  • error-tracking-audit

    Audits an app's error-tracking wiring against spec/project/error-tracking/ and produces a severity-classified audit artifact plus a plan for the gaps. The default `audit` operation dispatches the read-only error-tracking-audit-scanner agent, then hard-fails on any static tool-contract violation (SDK

  • guard-coverage-check

    Answers which rules a project claims, which of them are mechanically enforced, in which lane, and which exist only in prose. Dispatches the read-only `guard-coverage-scanner` agent to inventory the assertions in CLAUDE.md, architecture docs, ADRs, NFRs and specs, then persists a per-rule report unde

  • kpi-derive

    Derives a business application's project-specific KPIs from its goals, requirement documents, and source code, writing a human-readable artifact to project/kpis/<slug>.md per spec/project/kpi-definition-process/. The `derive` operation dispatches the read-only kpi-signal-scanner agent, walks a GQM r

  • license-check

    Runs an end-to-end license-compliance check on the current project per spec/project/license-check/ and produce a license-check audit artifact. Dispatches license-check-scanner agent for the read-only inventory (SBOM with resolved licenses, SPDX identification, category classification), then applies

  • maturity-assess

    Assesses the build maturity of an application's capabilities per spec/project/capability-maturity-assessment/, writing project/maturity/<slug>.md to surface where implementation quality is lacking. The `assess` operation inventories business-facing capabilities top-down (never bottom-up from directo

  • observability-audit

    Audits an app's observability against spec/project/monitoring-observability/ and produces a severity-classified audit artifact plus an implementation plan for the gaps. The default `audit` operation dispatches the read-only observability-audit-scanner agent, then hard-fails on any missing-or-unwired

  • quality-gate

    Runs the project's lint + typecheck + test gate in parallel, tabulates the results, and calls out exactly which checks failed so the caller can triage before a commit, a PR, or a release. Prefers repository-declared Taskfile targets (`task lint`, `task test`, `task typecheck`, `task check`) when the

  • release-regression-scope

    Determines the release-relevant regression/E2E test scope from a release change-set per spec/project/release-regression-scope/, so a team runs targeted-but-safe regression before rollout. The default `scope` operation resolves the release range, dispatches the read-only release-regression-scope-scan

  • source-code-review

    Runs a holistic senior-engineer source-code review of production and test code against spec/project/source-code-review/ plus its frontend extension spec/frontend/source-code-review/, and persists a severity-classified report whose disjoint work packages are parallel-dispatchable to specialists. The

  • test-cycle-orchestrate

    Drives one or more turns of the iterative test cycle defined in spec/project/test-cycle-foundation/ — case determination → execution → result analysis → code adaptation → re-execute — dispatching each phase to its capability (test-case-extractor and the per-tier test-generators for cases, quality-ga

  • test-pyramid-check

    Audits a feature's or module's test-tier completeness against the closed functional-tier taxonomy in spec/project/test-pyramid-foundation/ (are the applicable tiers — unit, component, integration, contract, E2E — present and written at the lowest tier that gives confidence?) and whether the E2E tier

  • webview-ui-optimize

    Audits a browser-rendered frontend against the canonical-language file under spec/frontend/webview-ui-optimization/ and, with per-item user approval, patches findings across five domains: Performance, Security, Accessibility (WCAG 2.2 AA), Internationalisation, and UX. Three operations: `audit` (wri

Subagents (39)

  • api-documentation-scanner

    Read-only scanner dispatched by the `api-documentation-audit` skill: detects the HTTP-API surface, discovers every OpenAPI document (conventional checked-in locations or a documented export command), then statically checks the version floor, info completeness, per-operation contract (operationId, ta

  • bdd-scenario-reviewer

    Read-only review of existing BDD scenarios (`.feature` files) and their step definitions per spec/project/behavior-driven-development/ and spec/project/bdd-page-object-integration/: declarative one-behavior-per-scenario Gherkin, intention-revealing titles, `@TC-<id>` traceability, no assertions in G

  • bjw-common-deployment-generator

    Senior deployment engineer that generates or extends a Helm chart on the bjw-s `common` library for a self-hosted app, applying spec/project/bjw-s-common-chart-deployment/ (structure) and spec/project/kubernetes-deployment-best-practices/ (hardened securityContext, default-deny NetworkPolicy, probes

  • capability-maturity-scanner

    Read-only scanner dispatched by the `maturity-assess` skill: given the skill's top-down capability inventory, mines the source tree and existing analysis reports for the machine-derivable signals of Axis B (static-analysis status, cyclomatic complexity, duplication, type coverage) and Axis C (per-ti

  • capability-reach-scanner

    Read-only scanner dispatched by the `capability-reach-audit` skill: given a local working copy, builds the audited set from the four declaration sources of spec/project/capability-reach-audit/ (requirement documents, specified endpoints such as workflow triggers and API contracts, documented capabil

  • code-security-reviewer

    Read-only, whole-codebase OWASP security audit that correlates findings across files (auth, access control, injection, secrets in logs, AI/RAG prompt-injection/SSRF) and returns a severity-classified report with file:line attribution and described (not applied) remediations. Invoke for a deep/full r

  • component-test-generator

    Scaffolds spec-conformant component-tier tests against spec/project/test-tier-component/ (a component in isolation with externals doubled at the boundary), in either flavour: frontend UI (assert observable output via user-facing queries) or service/backend (drive through its own API). Invoke to gene

  • component-test-reviewer

    Reviews existing component tests against spec/project/test-tier-component/ (frontend UI or service/backend), returns a checklist conformance verdict, and applies only minimal surgical fixes. Invoke to review, audit, or repair component tests. Don't use to scaffold them (`component-test-generator`),

  • contract-test-generator

    Scaffolds spec-conformant contract tests against spec/project/test-tier-contract/, verifying a service-to-service agreement without standing up both sides live. Defaults to consumer-driven (consumer records a contract; provider verified by replay), also provider-driven or bi-directional; asserts com

  • contract-test-reviewer

    Reviews existing contract tests against spec/project/test-tier-contract/, returns a checklist conformance verdict, and applies only minimal surgical fixes. Invoke to review, audit, or repair contract tests. Don't use to scaffold them (`contract-test-generator`), for another tier reviewer, or to audi

  • dependency-audit-scanner

    Read-only scanner dispatched by the `dependency-audit` skill: detects project type from lockfiles, runs the matching auditor (pip-audit, npm/pnpm/yarn audit, govulncheck, or cargo audit), and returns a structured per-package CVE drift inventory with severity and fixed-in version, per spec/project/de

  • deployment-bestpractices-reviewer

    Read-only auditor that reviews a Helm chart or Kubernetes deployment against spec/project/kubernetes-deployment-best-practices/ and returns a severity-classified checklist verdict with file:line attribution, covering the security and scalability pillars (NetworkPolicy, hardened securityContext, reso

  • deployment-change-analyzer

    Read-only analyzer that detects what changed in a self-hosted app relative to its existing bjw-s `common` Helm chart, so the chart can be extended to match. Scans the app source for runtime inputs the chart doesn't yet carry (new env vars, ports, services, persistent paths, secrets, egress, or a cha

  • dockerfile-audit-scanner

    Read-only scanner dispatched by the `dockerfile-audit` skill: discovers every Dockerfile and statically checks the mandatory OCI labels, the four mandatory non-label pillars (non-root numeric USER, no secrets in layers, base pinned by tag+digest, .dockerignore present), the advisory pillars, and (fo

  • e2e-result-reviewer

    Visually reviews an end-to-end run's outputs (screenshots and the machine-generated protocol) against the requirement, test-case, and UI specs, per spec/project/e2e-test-automation/. Reads each screenshot as an image and returns prioritised findings keyed to requirement/TC IDs. Read-only. Invoke aft

  • e2e-test-generator

    Scaffolds a spec-conformant end-to-end suite for a feature against spec/project/e2e-test-automation/ and the stability rules of spec/project/e2e-test-stability/ (self-provisioned test data, parallel-safety, guarded interaction helpers, truthful waits), defaulting to the Selenium + pytest reference p

  • e2e-test-reviewer

    Reviews an existing end-to-end suite against spec/project/e2e-test-automation/ and the stability rules of spec/project/e2e-test-stability/ (data isolation, parallel-safety, interaction hazards, skip/xfail hygiene) — Selenium + pytest reference profile — returns a checklist conformance verdict, and a

  • error-tracking-audit-scanner

    Read-only scanner dispatched by the `error-tracking-audit` skill: detects, per component, the static presence and wiring of the error-tracking tool contract per spec/project/error-tracking/ — SDK init at process entry, DSN from deployment config with a graceful no-DSN no-op, `environment` and `relea

  • frontend-code-reviewer

    Read-only senior-frontend review of components and their tests per spec/frontend/source-code-review/ (F1–F11 overlay) on top of spec/project/source-code-review/ (D1–D11): business logic leaking into the client, frontend error handling, component design, rendering and effects, styling and design-syst

  • frontend-usability-optimizer

    Senior UX engineer that improves the usability of already-implemented frontend code in place (forms, dialogs, list/table views, detail pages) against the project's own framework, component library, and UI conventions detected at runtime. Edits the presentation layer directly. Invoke to make existing

  • fullstack-developer

    Senior full-stack engineer that turns a sharply-scoped requirement into production-ready, runnable code (no pseudocode or stubs) against the consuming project's own tech stack, layout, and quality bar, detected at runtime. Implements features end-to-end across backend, frontend, and infrastructure,

  • gdpr-data-protection-reviewer

    Read-only, whole-repository GDPR/DSGVO data-protection audit. Discovers personal-data surfaces (data models, logging, third-party SDKs, cloud regions, frontend tags), detects personal-data classes incl. Art. 9, and audits against the GDPR principles, lawful basis, data-subject rights, privacy by des

  • guard-coverage-scanner

    Read-only scanner dispatched by `guard-coverage-check`: inventories the rules a repository asserts about itself (CLAUDE.md, architecture docs, ADRs, NFRs, specs) and returns a per-rule verdict of enforced, advisory, or prose-only, plus a separate drift flag where the text claims enforcement the repo

  • i18n-completeness-checker

    Read-only audit of a localized app's translation files for completeness against each other and against code usage. Diffs each locale against a reference, scans source for key references, and reports parity gaps, keys used but undefined (critical), and orphan keys. Adapts to the project's i18n librar

  • implementation-plan-author

    Given a GitHub issue id plus its requirements-elicit artifact, a confirmed requirements-elicit artifact with no issue, or an observability-audit or source-code-review findings report, authors an implementation plan: an atomic, testable work-package decomposition, each package naming the specialist a

  • integration-test-generator

    Scaffolds spec-conformant narrow integration tests against spec/project/test-tier-integration/ (the code against exactly one real external collaborator while doubling every other), asserting only the integration seam, with real-but-ephemeral dependencies (a disposable container, never a drifting in-

  • integration-test-reviewer

    Reviews existing integration tests against spec/project/test-tier-integration/, returns a checklist conformance verdict, and applies only minimal surgical fixes. Invoke to review, audit, or repair integration tests. Don't use to scaffold them (`integration-test-generator`), for another tier reviewer

  • kpi-signal-scanner

    Read-only scanner dispatched by the `kpi-derive` skill: mines the source tree and the requirement/goal documents for candidate KPI signals and returns a structured candidate inventory, each keyed to the goal/requirement it ladders back to, with file:line. Detection only: never selects the key KPIs,

  • license-check-scanner

    Read-only license-inventory scanner dispatched by the `license-check` skill: detects the stack (Python/Node/Go), reads the CycloneDX SBOM (or resolves licenses from lockfiles and registry metadata), maps each component to a canonical SPDX id and category, and returns the inventory plus the project's

  • observability-audit-scanner

    Read-only scanner dispatched by the `observability-audit` skill: detects static presence and wiring of the four mandatory observability pillars (metrics, structured logs with trace context, W3C Trace Context propagation, health/SLO/burn-rate alert) across backend, frontend floor, and third-party flo

  • python-code-reviewer

    Read-only senior-Python review of production and test code per spec/project/source-code-review/ (core dimensions + Python profile): correctness, design, domain-knowledge duplication, idiomatic usage, test health. Skips what ruff/mypy report; returns severity-classified findings with file:line and di

  • quality-gate-enforcer

    Read-only review of the quality-gate wiring (Taskfile targets, .pre-commit-config.yaml, ci.yml, timeouts) for spec-conformance against spec/project/quality-gate/, whether every tier the gate runs has an enforced lane among the required status checks, plus delimitation against workflow-health, depend

  • release-regression-scope-scanner

    Read-only scanner dispatched by the `release-regression-scope` skill: given a release change-set, attributes each change to its impacted topic area(s) by inverting the existing test-to-requirement traceability, and flags every change it can't mechanically attribute for the skill's worst-case fallbac

  • test-case-extractor

    Derives structured, framework-agnostic test cases from a requirement or specification document, decomposing it into user-observable-behaviour cases (no internal detail) as structured Markdown with full traceability and a coverage summary. Works in the source language and adapts to the project's inte

  • test-code-adapter

    Given a confirmed real failure (from `test-result-analyzer`), determines and applies the minimal correct production-code change that makes the red case pass, against spec/project/test-cycle-code-adaptation/, then re-runs to verify. Fixes the root cause and honours the no-cheating invariant (never we

  • test-result-analyzer

    Analyses the raw results of a test run against spec/project/test-cycle-result-analysis/ and classifies each non-pass (real defect / flake / test bug / infrastructure / stale dependency / config-secret drift) so the cycle knows what to do next, establishing flake-vs-real from the execution-emitted pe

  • unit-test-generator

    Scaffolds spec-conformant unit tests for a module or feature against spec/project/test-tier-unit/, defaulting to a pytest reference profile, with the FIRST properties, Arrange-Act-Assert, observable-behaviour assertions, and TC-ID traceability. Invoke to generate unit tests or turn test cases into r

  • unit-test-reviewer

    Reviews existing unit tests against spec/project/test-tier-unit/, returns a checklist conformance verdict, and applies only minimal surgical fixes. Invoke to review, audit, or repair unit tests. Don't use to scaffold them (`unit-test-generator`), for another tier reviewer, or to audit pyramid shape

  • webview-ui-expert

    Read-only, cross-file deep review of one named target against spec/frontend/webview-ui-optimization/ across five domains (Performance, Security, Accessibility WCAG 2.2 AA, i18n, UX). Produces a severity-sorted findings report with sources under spec/frontend/webview-ui-optimization/research/. Invoke

View source on GitHub nolte.github.io

More from nolte/claude-shared

All 5 plugins →

Shared Claude Code skills and agents for consistent development workflows across projects.

36 skills · 18 agents

0

Project-planning cadence for repositories that plan in project/: SMART mission authoring, roadmap init/planning/refinement, roadmap-item decomposition into features, and the sprint lifecycle from planning through execution to review, with the matching readiness, coherence, and consistency reviewers. Adopt this on top of nolte-shared only if your repository runs a mission/roadmap/feature/sprint cadence; repositories that track work as issues never do.

9 skills · 3 agents

0

Claude Code skill and agent authoring for plugin/skill developers: spec-conformant skill/agent scaffolding and revision, skill and agent review, portfolio-wide skills/agents sweeps, and skills/agents documentation-catalog generation. Adopt this on top of nolte-shared only if you author Claude Code skills or agents; the majority of consumers never do.

5 skills · 1 agent

0

Brand-aware image generation and media processing: provider-pluggable text-to-image, manual Gemini handoff, brand-conformant prompt authoring, and PNG-to-transparent-SVG vectorisation. Requires external image-generation credentials and binaries.

2 skills · 2 agents

0

Maintain nolte-engineering?

This page updates itself from your repo. Add the badge to your README, or get it Featured to pin it above the organic results with a followed link to your site.

claude & codex plugins5 plugins