By manikumarkv Verified
Modular SDLC plugin for any tech stack. Pick your tools, get tailored coding standards, MCP wiring, sub-agent context management, 35+ safety/quality hooks, and the full SDLC workflow — brainstorm to production.
Thorough code review of a branch against all standards. Generates REVIEW-<branch>.md with BLOCKER / WARNING / SUGGESTION findings. Trigger — "review this branch", "review my code", "check code quality", "pre-PR review".
Use when investigating bugs, errors, crashes, or unexpected behaviour. Also use for monitoring CloudWatch logs after a deployment or checking production health. Trigger phrases — "something is broken", "getting an error", "why is this failing", "check the logs", "monitor logs after deploy", "what's happening in production", "debug this", "root cause", "post-deploy check".
Runs eval cases from .eval.yaml files. For each case — loads skill context, generates code from the scenario prompt, checks must_contain and must_not_contain assertions, writes EVAL-<date>-<skill>.md reports.
Loads a single layer's detail file and answers a focused question about its rules. Returns a concise 5–15 line summary. Used by stack-dispatcher to keep main thread context clean.
Security-focused code review. OWASP Top 10, Cognito JWT patterns, secrets scanning, IAM permissions, PII in logs. Outputs SECURITY-REVIEW-<branch>.md. Trigger — "security review", "check for vulnerabilities", "OWASP check", "scan for secrets".
Web accessibility (a11y) standards — semantic HTML, ARIA, keyboard navigation, focus management, forms, dynamic content. Load when writing or reviewing any UI component or page.
Create a numbered Architecture Decision Record in docs/adr/ using the Nygard format. Links the ADR back to the dev-brainstorm doc if one exists. Usage — /adr <title> [issue-number]
Universal REST API design principles — response envelope, route naming, versioning, pagination, status codes. Applies to any backend language or framework. Load when designing or implementing any API endpoint.
Manage git branches and scaffold new feature boilerplate. Usage — /devrunway:branch <sub-command> [args]
Universal quality checklists for every development action — feature addition, API endpoint, data model change, logging, secrets, auth. Auto-apply the relevant checklist whenever one of these actions is taken.
Executes bash commands
Hook triggers when Bash tool is used
Modifies files
Hook triggers on file write and edit operations
Requires secrets
Based on adoption, maintenance, documentation, and repository signals. Not a security audit or endorsement.
This plugin requires configuration values that are prompted when the plugin is enabled. Sensitive values are stored in your system keychain.
JIRA_EMAILJira account email. Required if project-management=jira.
${user_config.JIRA_EMAIL}FIGMA_TOKENFigma Personal Access Token (from Figma → Settings → Personal Access Tokens). Used for reading design files and components.
${user_config.FIGMA_TOKEN}GITHUB_TOKENGitHub fine-grained PAT. Required scopes: Contents (read), Issues (read/write), Pull Requests (read/write), Metadata (read). Generate at github.com → Settings → Developer settings → Personal access tokens.
${user_config.GITHUB_TOKEN}GITLAB_TOKENGitLab PAT. Required scopes: api, read_repository. Required if project-management=gitlab or source-control=gitlab.
${user_config.GITLAB_TOKEN}JIRA_BASE_URLJira Cloud base URL (e.g. https://yourorg.atlassian.net). Required if project-management=jira.
${user_config.JIRA_BASE_URL}GITLAB_API_URLGitLab API URL (default https://gitlab.com/api/v4 for SaaS; set to self-hosted instance otherwise).
${user_config.GITLAB_API_URL}JIRA_API_TOKENJira API token (Atlassian → Account → Security → API tokens). Required if project-management=jira.
${user_config.JIRA_API_TOKEN}LINEAR_API_KEYLinear API key (Linear → Settings → API → Personal API keys). Required if project-management=linear.
${user_config.LINEAR_API_KEY}NOTION_API_KEYNotion internal integration token (Notion → Settings → Integrations → New integration). Required if documents=notion.
${user_config.NOTION_API_KEY}CONFLUENCE_EMAILAtlassian account email. Required if documents=confluence.
${user_config.CONFLUENCE_EMAIL}CONFLUENCE_BASE_URLConfluence Cloud base URL (e.g. https://yourorg.atlassian.net/wiki). Required if documents=confluence.
${user_config.CONFLUENCE_BASE_URL}CONFLUENCE_API_TOKENConfluence API token (Atlassian → Account → Security → API tokens). Required if documents=confluence.
${user_config.CONFLUENCE_API_TOKEN}Needs API keys or credentials to function
Needs API keys or credentials to function
Uses power tools
Uses Bash, Write, or Edit tools
Uses power tools
Uses Bash, Write, or Edit tools
Extend Claude Code with a full SDLC skill set for any tech stack — configure once, works everywhere.
devrunway is a Claude Code plugin that adds:
Inside Claude Code:
# 1. Add the devrunway marketplace
/plugin marketplace add manikumarkv/devrunway-claude-plugin
# 2. Install the plugin
/plugin install devrunway@devrunway
# 3. Run the setup wizard to configure your stack
/setup
/setup asks 35 questions across 6 screens and generates:
| Output | What it does |
|---|---|
stack.json | Declares which tech layers you use |
.mcp.json | Pre-configures MCP servers (Figma, GitHub, Jira…) |
All 135 layer skills are bundled with the plugin. There is no per-layer install step. The stack-dispatcher agent reads stack.json and the files you edit, then loads only the relevant layer detail files into a sub-agent so your main thread stays light.
Your stack: React + Node/Express + AWS + PostgreSQL/Prisma + Cognito
shadcn + Tailwind + Zustand + Zod
Pino → CloudWatch + Sentry
Vitest + Playwright + Bruno + MSW
Figma + GitHub + GitHub Actions
Layers activated for your stack (auto-load on matching files):
layers/source-control/github
layers/ci/github-actions
layers/frontend/react
layers/css/tailwind
layers/ui-components/shadcn
layers/state/zustand
layers/backend/node-express
layers/validation/zod
layers/cloud/aws
layers/database/postgres-prisma
layers/auth/cognito
layers/logging/framework/pino
layers/logging/provider/cloudwatch
layers/error-monitoring/sentry
layers/testing/unit/vitest
layers/testing/e2e/playwright
layers/testing/api/bruno
layers/mocking/msw
layers/design/figma
MCP servers configured in .mcp.json (auto-registered on plugin install):
figma → @figma/mcp-server (FIGMA_ACCESS_TOKEN)
github → @modelcontextprotocol/server-github (GITHUB_PERSONAL_ACCESS_TOKEN)
skills/ ← universal slash commands (product-plan, dev-code, pr, eval, forge, friction…)
agents/ ← code-reviewer, security-reviewer, debugger, stack-dispatcher, layer-consultant
hooks/ ← 35 active safety + quality hooks (destructive-git-guard, secrets-leak-guard, …)
layers/ ← 135 technology layers, auto-loaded by file pattern
frontend/ react | vue | angular | nextjs
backend/ node-express | python-fastapi | python-django | dotnet
cloud/ aws | gcp | azure
database/ postgres-prisma | dynamodb | mongodb | sqlalchemy
auth/ cognito | firebase | auth0 | azure-ad
css/ tailwind | styled-components | css-modules | bootstrap
ui-components/ shadcn | mui | ant-design | chakra
state/ zustand | redux-toolkit | jotai | pinia
validation/ zod | yup | valibot | joi
testing/
unit/ vitest | jest | pytest
e2e/ playwright | cypress | selenium
api/ bruno | postman | insomnia
mocking/ msw | mirage | json-server
logging/
framework/ pino | winston | morgan
provider/ cloudwatch | datadog | splunk | newrelic
error-monitoring/ sentry | datadog-apm | bugsnag
ci/ github-actions | gitlab-ci | circleci | azure-pipelines
design/ figma | sketch | adobe-xd
payment/ stripe | paypal | braintree
storage/ s3 | cloudinary | gcs | uploadthing
search/ algolia | elasticsearch | typesense
realtime/ socketio | pusher | ably
cache-queue/ redis | bullmq | sqs | rabbitmq
feature-flags/ launchdarkly | aws-appconfig | flagsmith | posthog
secrets/ aws-secrets-manager | vault | doppler | env-only
source-control/ github | gitlab | bitbucket | azure-devops
... and more (135 layers total)
setup/ ← /setup wizard
/product-brainstorm → /product-plan → /product-tasks → /product-refine →
/dev-brainstorm → /dev-design → /dev-code → /dev-review → /pr create →
/deploy staging → /validate → /deploy prod
npx claudepluginhub manikumarkv/devrunway-claude-plugin --plugin devrunwayHarness-native ECC operator layer - 67 agents, 271 skills, 92 legacy command shims, reusable hooks, rules, selective install profiles, and production-ready workflows for Claude Code, Codex, OpenCode, Cursor, and related agent harnesses
Comprehensive feature development workflow with specialized agents for codebase exploration, architecture design, and quality review
Reliable automation, in-depth debugging, and performance analysis in Chrome using Chrome DevTools and Puppeteer
Access thousands of AI prompts and skills directly in your AI coding assistant. Search prompts, discover skills, save your own, and improve prompts with AI.
Comprehensive skill pack with 66 specialized skills for full-stack developers: 12 language experts (Python, TypeScript, Go, Rust, C++, Swift, Kotlin, C#, PHP, Java, SQL, JavaScript), 10 backend frameworks, 6 frontend/mobile, plus infrastructure, DevOps, security, and testing. Features progressive disclosure architecture for 50% faster loading.
Tools to maintain and improve CLAUDE.md files - audit quality, capture session learnings, and keep project memory current.