cure-product-engineering

Plugin cure-product-engineering from marketplace cure in Cure-Consulting-Group/ProductEngineeringSkills. Commands read from the repo's manifest — today.

Skills (104)

• bid-decision

  Makes the bid/no-bid call: kill criteria, scorecard, win odds, bid cost. Use when deciding whether to commit to a proposal, or when a pursuit is drifting.

• burn-rate-tracker

  Models burn, runway scenarios, break-even, and cash-flow projections. Use when asking how long cash lasts, when to raise or cut, or planning a studio or product budget.

• buyer-intelligence

  Builds a buyer, incumbent, and renewal-date database from lost bids. Use when a bid is declined, a sole source or award notice appears, or building a buyer dossier.

• capture-management

  Builds a public-sector bid pipeline: targets, sourcing, teaming. Use when entering a government market, choosing which bids to chase, or building past performance.

• engineering-cost-model

  Estimates internal build cost: dev hours, cloud, maintenance, build-vs-buy. Use when asking what a feature or product will cost or take to build.

• finops

  Cuts Firebase and GCP cost: budgets, alerts, right-sizing, AI API spend. Use when a cloud bill spikes, budgeting a live bill, or cutting infra and LLM costs.

• fundraising-materials

  Builds pitch decks, the ask, use of funds, and investor outreach pipelines. Use when preparing a seed or Series A raise: deck, intro blurb, target list, or process plan.

• investor-reporting

  Drafts investor updates, board decks, portfolio P&L, data rooms, cap tables. Use when writing the monthly investor update, a board deck, a SAFE or cap-table model, or diligence prep.

• proposal-generator

  Drafts consulting proposals and SOWs: scope, milestones, pricing, terms. Use when writing a client proposal, a statement of work, or a change order for Cure.

• public-sector-contracting

  Reviews government contract terms: liability, IP, non-appropriation, exit. Use when reading a municipal, state, or federal contract form or its insurance terms, or choosing exceptions.

• rfp-evaluation

  Extracts every RFP/RFQ requirement into a compliance matrix. Use when a solicitation has passed triage and you need what it requires and how its rubric maps to effort.

• saas-financial-model

  Models SaaS unit economics, MRR/ARR projections, and pricing tiers. Use when computing LTV, CAC, payback, or churn, projecting MRR, or setting subscription prices.

• solicitation-triage

  Screens public-sector solicitations and portal alerts to a fast verdict. Use when a new RFP, RFQ, bid notice, or weekly bid-portal feed arrives.

• technical-estimation

  Builds defensible software estimates: PERT, reference class, risk reserve. Use when an estimate will be bid on, contracted, or audited, e.g. an RFP cost volume.

• agent-designer

  Designs LLM agent internals: tool schemas, memory, termination, evals. Use when an agent or multi-agent system is the chosen shape and needs a design.

• agent-harness

  Use when a repo's agents break prose rules or finish on red. Installs Cure's harness kit: AGENTS.md map with Controls, a sub-60s verify gate, Stop hook, ask/deny permissions, mutation-checked contract tests.

• agent-workflow-designer

  Picks the shape of an LLM system: single call, workflow pattern, or agent. Use when deciding how to structure an AI feature, or when an agent is too slow, costly, or flaky.

• ai-feature-builder

  Builds LLM features: client wrapper, prompts, streaming UX, kill switch. Use when adding chat, summarization, extraction, or generation to an app, with fallbacks.

• analytics-implementation

  Designs analytics event taxonomies, tracking plans, funnels, consent flows. Use when instrumenting product analytics (GA4/Firebase, Mixpanel, PostHog) or auditing events.

• android-design-expert

  Applies Material Design 3 to Android screens and components. Use when designing or reviewing Compose UI for dynamic color, adaptive layouts, navigation, motion, or M3 Expressive.

• android-feature-scaffold

  Scaffolds Android feature modules (Clean Architecture, MVI, Compose, Hilt). Use when creating a new Kotlin feature, screen, ViewModel, repository, or use case in an Android app.

• api-architect

  Designs REST/GraphQL API contracts: errors, auth, rate limits, versioning. Use when designing endpoints, an OpenAPI spec, or an API's error, auth, or deprecation policy.

• api-gateway

  Designs API gateway and BFF layers: middleware, aggregation, federation. Use when adding a gateway, a mobile/web BFF, or federating GraphQL subgraphs.

• client-communication

  Drafts client status emails, escalations, exec summaries, and demo scripts. Use when writing a weekly client update, escalating a blocker, or preparing a stakeholder demo or QBR.

• client-handoff

  Builds client handoff packages: runbooks, credential transfer, KT, SLA. Use when handing a project, phase, or support role to a client team or winding down an engagement.

• cure-infra-bootstrap

  Runs Cure's bootstrap CLI (init/apply/doctor/inventory). Use when setting up, upgrading, or drift-checking CLAUDE.md, .claude/, and rules in a project.

• data-migration

  Plans data migrations: ETL, backfills, dual-write, cutover, rollback. Use when moving or reshaping existing data across databases, Firestore, or legacy systems without downtime.

• database-architect

  Chooses databases and designs SQL/Room schemas, indexes, and query plans. Use when picking an engine, modeling relational data, adding indexes, or fixing slow queries.

• e2e-testing

  Writes E2E suites (Playwright, Compose, XCUITest): page objects, CI wiring. Use when adding E2E, smoke, or visual regression tests for a user flow, or fixing flaky E2E tests.

• env-secrets-manager

  Audits .env files and secrets, read-only. Use when designing an .env schema, scanning for leaked keys, responding to a leak, planning rotation, or moving to a secret manager.

• firebase-architect

  Designs Firestore models, security rules, Cloud Functions v2, and App Check. Use when building a Firebase feature, writing rules or triggers, or wiring a Firestore data layer.

• git-worktree-manager

  Sets up git worktrees with isolated ports, env files, and databases. Use when running parallel features, a hotfix, or a PR review without stashing or switching the main checkout.

• i18n

  Internationalizes and localizes Android, iOS, and Next.js apps. Use when externalizing strings, adding a locale or RTL, fixing plurals or date/currency formats, or setting up translation.

• interview-system-designer

  Designs engineering interview loops: stages, rubrics, question banks. Use when hiring engineers, setting an AI-use policy or debrief rules, or fixing a loop with poor pass or accept rates.

• ios-architect

  Scaffolds SwiftUI features: Clean Architecture, MVVM, Swift 6 concurrency. Use when creating a new iOS feature, view model, repository, or StoreKit 2 flow.

• ios-design-expert

  Applies Apple HIG to iOS/iPadOS screens and components. Use when designing or reviewing SwiftUI UI for Liquid Glass, navigation, tab bars, Dynamic Type, SF Symbols, or widgets.

• llmops

  Runs LLM features in production: evals, prompt versions, cost, routing. Use when an AI feature needs eval gates, spend control, or a model or prompt rollout.

• mcp-server-builder

  Designs and builds MCP servers: tool schemas, resources, transports, auth. Use when building a Model Context Protocol server, wrapping an API as MCP tools, or migrating to a new SDK/spec.

• micro-frontends

  Decides whether and how to split a web frontend into independent apps. Use when teams block each other's releases, or weighing Vercel microfrontends, multi-zones, or module federation.

• monorepo-navigator
• nextjs-feature-scaffold
• notification-architect
• offline-first
• parallel-agent-orchestration
• performance-review
• project-bootstrap
• project-manager
• rag-architect
• sdlc
• self-improving-memory
• stitch-design
• stripe-integration
• test-accounts
• testing-strategy
• web-design-expert
• valuation
• bar-benchmark
• legal-doc-scaffold
• legal-doctrine
• legal-research
• go-to-market
• growth-engineering
• instagram-publishing-setup
• product-marketing
• seo-content-engine
• technical-blog-writer
• chaos-engineering
• ci-cd-pipeline
• disaster-recovery
• dora-metrics
• edge-computing
• engagement-automation
• green-software
• incident-response
• infrastructure-scaffold
• observability
• release-management
• customer-onboarding
• design-studio
• design-system
• feature-audit
• feature-flags
• market-research
• portfolio-registry
• product-design
• product-manager
• technology-radar
• uat
• accessibility-audit
• compliance-architect
• qsbs-compliance
• security-review
• audit-risk-substantiation
• cpa-benchmark
• cpa-standards
• deductions-and-credits
• estimated-tax-compliance
• irc-lookup
• nonprofit-dissolution
• return-review
• software-dev-tax
• tax-preparation
• tax-recommendations
• tax-strategies

Subagents (41)

• ab-test-analyst

  Designs and analyzes A/B tests: sample size, significance, guardrails. Use when designing an experiment, sizing a test, or interpreting A/B results.

• accessibility-checker

  WCAG 2.2 AA check of web, Android, and iOS UI code. Use when UI changes need an accessibility pass; reports every finding with severity and confidence.

• api-validator

  Checks API code against its OpenAPI or GraphQL spec. Use when validating contract coverage, error consistency, or breaking changes before merge or release.

• brand-guardian

  Audits voice, terminology, microcopy, and visual tokens for consistency. Use when reviewing UI copy, naming, or visual identity against a brand or style guide.

• campaign-analyst

  Analyzes marketing attribution, funnels, CAC/LTV, and channel ROI. Use when evaluating campaign performance or auditing UTM and conversion tracking.

• ci-debugger

  Diagnoses failed CI/CD runs from logs and proposes the minimal fix. Use when a GitHub Actions, Firebase, Fastlane, or Docker build fails.

• code-reviewer

  Reviews named files or uncommitted work against Cure standards. Use when checking work in progress before a commit; every finding with severity and confidence, no merge verdict. NOT for a branch before merge (pr-reviewer agent) or a bug-only pass (/code-review).

• codebase-explainer

  Explains a codebase with file:line citations. Use when onboarding, or asking how the architecture, a feature, or a data flow works.

• competitive-intel

  Competitive analysis: feature matrices, positioning, pricing, moats. Use when comparing the product to competitors or looking for differentiation gaps.

• content-strategist

  Content strategy: editorial calendars, briefs, SEO pillars, distribution. Use when planning content or auditing existing content against growth goals.

• contract-reviewer

  Business-risk review of SOWs, MSAs, NDAs, and contracts. Use when checking terms for scope, payment, IP, liability, or termination risk. Not legal advice.

• data-analyst

  Explores schemas, writes analytical queries, and checks data quality. Use when mapping a data model, answering a data question, or investigating anomalies.

• dependency-auditor

  Audits dependencies for CVEs, staleness, licenses, and supply-chain risk. Use after adding or upgrading packages, or before a release.

• deployment-validator

  Pre-deploy gate: env vars, secrets, flags, build, tests, rollback readiness. Use before deploying to staging or production; returns GO/NO-GO with findings.

• doc-generator

  Generates docs from source: API docs, ADRs, changelogs, onboarding guides. Use when documentation is missing or has drifted from the code.

• equity-analyst

  Public-equity research: filings, earnings, catalysts, valuation thesis. Use when researching a listed company or updating a thesis after earnings.

• financial-analyst

  Builds forecasts, unit economics, and scenarios from pricing and cost code. Use when you need a revenue forecast, unit-economics model, or P&L projection.

• firebase-security-auditor

  Audits Firestore, Storage, and RTDB security rules against the data model. Use after changing rules or Firestore schema, or before a Firebase deploy.

• growth-analyst

  Analyzes activation, retention, virality, and monetization. Use when looking for growth levers or diagnosing a funnel or retention drop from code and data.

• investment-banker

  M&A and capital-markets analysis: comps, DCF, LBO, accretion/dilution. Use when valuing a company for a deal or drafting a teaser, CIM, or buyer list.

• investor-relations

  Drafts board updates, investor reports, KPI packs, and raise narratives. Use when preparing investor-facing materials from product and financial data.

• legal-analyst

  Drafts NY-first legal research memos with every citation verified, for attorney review. Use when a legal question needs an IRAC answer with authority, or a draft's citations need checking.

• legal-compliance

  Flags QSBS, FERPA, NCAA, trademark, and entity-compliance risk. Use when checking a portfolio company or codebase against those regimes. Not legal advice.

• market-intelligence

  Market sizing (TAM/SAM/SOM), industry structure, trends, and timing. Use when validating a market, sizing an opportunity, or testing an investment thesis.

• metrics-dashboard

  Specifies KPIs, SLOs/SLIs, alert thresholds, and dashboard layouts. Use when designing dashboards or alerting for engineering, product, or business metrics.

• migration-validator

  Reviews DB migrations for rollback safety, locking, and zero-downtime fit. Use before applying migrations to staging or production; reports every finding.

• ops-finance

  Operational finance: invoices, 1099 tracking, month-end close, filing inventory. Use for Cure bookkeeping and multi-entity ops; tax positions go to tax-analyst.

• pr-reviewer

  Reviews a branch against its base for bugs, security, performance, tests, standards. Use before merging a PR; every finding with severity and confidence, plus a merge verdict. NOT for uncommitted work or a file set (code-reviewer agent).

• private-equity-analyst

  Private-markets analysis: screening, diligence, LBO returns, IC memos. Use when sourcing or diligencing a private deal or monitoring a portfolio company.

• product-analyst

  Analyzes feature adoption, user journeys, and PMF signals. Use when auditing analytics coverage or asking how users move through the product.

• project-bootstrapper

  Scaffolds a new Android, iOS, Next.js, or Firebase project to Cure standards. Use when starting a repo from scratch, not when adding features to an existing one.

• qa-engineer

  Adversarial QA: test plans, edge cases, regression scope, bug triage. Use when a change needs a QA pass beyond running tests; gives a QA ship recommendation that feeds deployment-validator's GO/NO-GO, never the deploy call itself.

• refactor-assistant

  Behavior-preserving refactors with tests run before and after each step. Use when restructuring code that must not change behavior; stops on a red baseline.

• release-coordinator

  Runs a release: version bump, changelog, validation, tag, rollback plan. Use when cutting a web, mobile, or backend release; confirms before tagging or publishing.

• roadmap-strategist

  Builds roadmaps with RICE scoring, dependencies, and capacity. Use when prioritizing a backlog, sequencing work, or planning the next quarter.

• skill-security-auditor

  Static security audit of skill, agent, and persona files. Use before merging changes under skills/, agents/, or personas/; returns PASS/WARN/FAIL and every finding.

• system-architect

  Architecture reviews, RFCs, and ADRs with explicit trade-offs. Use when choosing a technology, drawing service boundaries, or reviewing a system design.

• tax-analyst

  Drafts tax workpapers, reviews, estimates, and plans for CPA review. Use when preparing or reviewing a return, planning estimates, scoring audit risk, or treating dev costs.

• technical-content-strategist

  Turns engineering work into plain-language posts and visuals. Use when writing a technical story for business readers, with analogies and ROI framing.

• test-runner

  Runs the test suite, checks coverage, and flags skipped or flaky tests. Use after writing code or before a commit; reports results and doesn't fix code.

• ux-researcher

  Usability review from UI code: IA, flows, forms, feedback, cognitive load. Use when auditing a product's UX or planning user research.

Hooks (11)

• SessionStart

  echo 'Cure Consulting Group ProductEngineeringSkills plugin loaded (v7.16.1). 104 skills (domain-organized), 41 agents, 4 personas, 11 rules, 9 output styles. Use /cure-product-engineering:<skill-name

• SessionStart

  python3 "${CLAUDE_PLUGIN_ROOT}/hooks/cure_guard.py" session-start

• SessionStart

  printf '%b\n' 'STANDARDS RE-INJECTED AFTER COMPACTION — Cure Consulting Group standards (always apply):\n- Clean Architecture: domain (pure logic) / data (DTOs + repos) / presentation (VMs + UI)\n- An

• UserPromptSubmit

  python3 "${CLAUDE_PLUGIN_ROOT}/hooks/cure_guard.py" prompt

• PreToolUse

  python3 "${CLAUDE_PLUGIN_ROOT}/hooks/cure_guard.py" edit

• PreToolUse

  python3 "${CLAUDE_PLUGIN_ROOT}/hooks/cure_guard.py" skill-content

• PreToolUse

  python3 "${CLAUDE_PLUGIN_ROOT}/hooks/cure_guard.py" bash

• PreToolUse

  python3 "${CLAUDE_PLUGIN_ROOT}/hooks/cure_guard.py" telemetry

• PostToolUse

  python3 "${CLAUDE_PLUGIN_ROOT}/hooks/cure_guard.py" agent-done

• PostToolUseFailure

  python3 "${CLAUDE_PLUGIN_ROOT}/hooks/cure_guard.py" failure

• Stop

  python3 "${CLAUDE_PLUGIN_ROOT}/hooks/cure_guard.py" stop

Language servers (2)

• typescript
• python

Similar plugins