solo-sdlc

Plugin solo-sdlc from marketplace shakhoff-claude-marketplace in shakhovskiya-create/shakhoff-claude-marketplace. Commands read from the repo's manifest — 1 month ago.

Skills (21)

• adr-writer

  Write Architecture Decision Records (ADRs) for any meaningful technical decision. Use this skill whenever an architectural choice is made (database, framework, protocol, library, pattern). Triggers when an architecture decision is being made, when alternatives are being weighed, or when the user ask

• architecture

  Design the system architecture from an approved PRD. Use this skill in Phase 3 of every feature, AFTER PRD is approved and BEFORE writing any code. Triggers on requests for system design, architecture, design doc, "how to build this", tech choices, or component design. Produces a design doc with C4

• cache-investigation

  Persist one-shot investigations (architecture survey, dependency map, library research) to docs/cache/<topic>/ so they're not re-done each session. Use BEFORE running any deep investigation skill (data-source-research, system-architect surveys). Reuse if cache fresh; otherwise refresh.

• code-discipline

  Apply LLM coding discipline to every edit - minimum code, surgical changes, surface assumptions, goal-driven execution. Use this skill ALWAYS when writing or editing code, especially when modifying existing files. Triggers on any code-writing task, refactor, bug fix, or feature implementation. Inspi

• code-review

  Multi-perspective code review (correctness, security, performance, maintainability) on implemented changes. Use this skill AFTER verification passes and BEFORE the work is shipped. Triggers when implementation is complete, before merging, or when user asks for review. Outputs severity-tagged finding

• continuous-improvement

  Generate concrete, actionable improvement recommendations after every phase, every implementation, and every review. Use this skill at the END of every Claude Code task. Triggers when a phase completes, after implementation is done, after a code review, or when the user asks "what could be better".

• data-source-research

  Investigate a data source (database, API, file system, message queue) ONCE and produce a permanent reference document. Use this skill BEFORE writing any code that reads from or writes to a data source. Triggers when the user mentions a database, API, table, schema, or any external data they want to

• ddd-modeling

  Apply Domain-Driven Design to identify bounded contexts, aggregates, entities, value objects, and the ubiquitous language. Use this skill during architecture phase when the domain has multiple subdomains, complex business rules, or non-trivial invariants. Triggers on mentions of DDD, bounded context

• design-md

  Maintain a DESIGN.md file (Google Labs format) describing the project's visual identity for coding agents. Use this skill whenever a project has a design system, when extracting tokens from a mockup, when frontend agents need to know design tokens, or when working with shared component libraries. Th

• documentation

  Generate and maintain technical documentation - README, API docs, runbooks, architecture diagrams, getting-started guides. Use when documentation is requested or when shipping a feature that needs docs. Triggers on requests for README, docs, API spec, runbook, getting-started, onboarding, or after a

• interview

  Interview the user before writing any code. Use this skill at the START of every new feature, project, or non-trivial task. ALWAYS use this when the user describes what they want in natural language without complete specs. Triggers on phrases like "I want to build", "let's create", "make me a", "I n

• lsp-setup

  Configure Language Server Protocol (LSP) for the project's languages so Claude Code can use real diagnostics, auto-completion, and refactoring tools. Use this skill when starting a new project or when joining an existing one without LSP configured. Triggers when user mentions LSP, language server, d

• mock-fidelity

  Implement a frontend that is pixel-equivalent to an approved mockup. Use this skill whenever the user provides a mockup, Figma design, screenshot, or design reference and wants the UI built to match. Triggers on phrases like "make it look like this mockup", "implement this design", "1:1 with the moc

• overnight-loop

  Set up Claude Code to safely run autonomously for hours (overnight, multi-hour batch work). Use this skill when the user wants to leave Claude working unattended on a well-defined batch of tasks. Triggers on phrases like "leave this overnight", "run while I sleep", "autonomous run", "process this ba

• prd-writer

  Write a Product Requirements Document (PRD) with functional and non-functional requirements. Use this skill in Phase 2 of every feature, AFTER the discovery brief is complete and BEFORE architecture. Triggers when the user has approved a brief and asks for the PRD, technical spec, ТЗ, requirements d

• sprint-planning

  Plan a sprint from a backlog of user stories. Use when stories exist and the user wants to organize them into a sprint, batch of work, or implementation plan. Triggers on requests for sprint plan, milestone plan, what to build first, or backlog organization. Outputs an ordered, dependency-respecting

• tdd

  Apply test-driven development with strict red-green-refactor cycles. Use this skill during implementation phase for any logic that has correctness requirements (business rules, calculations, state transitions, data transformations, parsers). Triggers when user asks for implementation, "write the cod

• user-stories

  Generate well-formed user stories with INVEST criteria from a PRD. Use when the PRD is approved and stories need to be created for sprint planning. Triggers on requests for user stories, agile stories, sprint backlog, or story splitting. Outputs stories that satisfy INVEST (Independent, Negotiable,

• verification-contract

  Emit machine-readable verification report from solo-sdlc:verify. Schema-validated against verification-report.schema.json. Consumed by ai-scrum:dod-validator and code-reviewer to avoid duplicate work.

• verify

  Verify that implementation actually works by running tests, running the app, and exercising it. Use this skill at the END of every implementation step BEFORE declaring anything done. Triggers when implementation is complete, when user asks "does it work", or before any "ready for review" claim. Bori

• wiki-workflow

  Manage a persistent, LLM-maintained knowledge wiki for accumulated knowledge that compounds over time. Use this skill when the user has a domain where knowledge accumulates over weeks/months (research, business context, infrastructure knowledge, personal notes), or when they ask about "wiki", "knowl

Slash commands (9)

• /solo-sdlc:audit-config

  Audit the .claude/ setup for conflicts, stale instructions, drift from current Claude Code spec.

• /solo-sdlc:babysit

  Watch open PRs, automatically respond to code review comments, keep branches rebased. Use with /loop for continuous PR maintenance.

• /solo-sdlc:commit-push-pr

  Commit current changes, push to a feature branch, open a PR with auto-generated description.

• /solo-sdlc:implement-mock

  Implement frontend pixel-equivalent to a provided mockup. Extracts tokens, builds components in isolation, verifies side-by-side.

• /solo-sdlc:investigate-db

  Investigate a data source ONCE and save reference to docs/data-sources/. Reuses existing reference if fresh.

• /solo-sdlc:new-feature

  Start a new feature from scratch - runs the full SDLC pipeline (Discovery → Requirements → Architecture → Implementation → Verification → Ship)

• /solo-sdlc:overnight

  Set up an overnight autonomous run with safety guardrails. User defines scope, this orchestrates the loop with rollbacks, per-task commits, status reports.

• /solo-sdlc:sprint-plan

  Plan a sprint from current backlog of user stories. Outputs ordered, dependency-respecting plan.

• /solo-sdlc:verify-app

  Run full verification on current changes - tests, app run, acceptance criteria, observability, performance, visual.

Subagents (13)

• backend-django

  MUST BE USED for Django implementation - models, views, DRF, admin, signals, migrations. Use when the project is a Django app or Django REST Framework API. Knows the idiomatic Django patterns and the common pitfalls.

• backend-go

  MUST BE USED for Go backend implementation. Builds Go services, libraries, CLIs following idiomatic Go and the project's CLAUDE.md rules. Use when implementing in Go - APIs, workers, CLIs, gRPC services.

• backend-python

  MUST BE USED for Python backend implementation (FastAPI, Flask, plain Python services). Use when implementing Python services, APIs, workers, CLIs that don't require Django specifically. For Django projects, prefer backend-django.

• code-reviewer

  MUST BE USED after verification passes and before commit/PR. Multi-perspective review: correctness, security, performance, maintainability. Severity-tagged findings (Critical / High / Medium / Low). Stack-aware (Go, Python, Django, React, Vue).

• data-investigator

  MUST BE USED before writing any code that touches a new data source (database, API, queue). Investigates the source ONCE and produces a permanent reference at docs/data-sources/. PROACTIVELY check for existing reference before re-investigating. Solves the "Claude keeps re-asking about the database"

• database-architect

  MUST BE USED for database schema design, migrations, indexing strategy, query optimization, partitioning. Use whenever a feature touches DB design or DB performance. Knows PostgreSQL, ClickHouse, MySQL well; can advise on others.

• debugger

  MUST BE USED when something breaks - tests fail, app crashes, unexpected behavior, performance regression. Applies systematic debugging: root cause analysis, not symptom-patching. Use PROACTIVELY when verification fails or when the user reports a bug.

• frontend-react

  MUST BE USED for React frontend implementation. Use for React 18+ apps, component libraries, Next.js apps. Implements UI from mockups using mock-fidelity skill, builds accessible components, handles state correctly.

• frontend-vue

  MUST BE USED for Vue 3 frontend implementation. Use for Vue 3 SFC apps, Nuxt 3, component libraries. Implements UI from mockups using mock-fidelity skill, follows Composition API patterns.

• mock-implementer

  MUST BE USED whenever the user provides a UI mockup, screenshot, or design and wants the frontend built to match. Builds pixel-equivalent UI by extracting tokens, building components in isolation, then verifying with side-by-side screenshots. Solves the "frontend looks nothing like the mockup" probl

• product-analyst

  MUST BE USED at the start of any new feature or project. Handles Phase 1 (Discovery) - asks the right questions, identifies users, scopes the work, and produces a discovery brief. Use PROACTIVELY when the user describes intent without complete specs.

• system-architect

  MUST BE USED in Phase 3 (Architecture) after the PRD is approved. Designs the system using DDD where appropriate, produces C4 diagrams, writes ADRs for major choices, runs the solutioning gate check. Use PROACTIVELY before any implementation begins.

• verifier

  MUST BE USED at the end of every implementation task BEFORE marking done. Runs tests, runs the app, exercises acceptance criteria, checks observability. Boris Cherny: "give the agent a way to verify and it improves the result." NEVER skip - this is what produces "first-try" quality.

Hooks (3)

• PreToolUse

  ${CLAUDE_PLUGIN_ROOT}/hooks/pre-tool-use.sh

• PostToolUse

  ${CLAUDE_PLUGIN_ROOT}/hooks/post-tool-use.sh

• SessionStart

  ${CLAUDE_PLUGIN_ROOT}/hooks/session-start.sh

Similar plugins