By everva
Dark Factory: Autonomous AI development pipeline with governance tiers, holdout validation, satisfaction judging, and Ralph Wiggum loop.
Evaluate project readiness for autonomous development. Scores 8 axes and produces a JSON scorecard.
Run test suites across the project to catch regressions. Use after significant changes or before merging major features.
Two-pass adversarial LLM evaluation of implementation quality across 5 dimensions.
Analyze incoming tasks: classify size (XS/S/M/L/XL), identify affected areas, detect dependencies, recommend pipeline depth.
Validates implementation against hidden behavioral scenarios that the implementation agent never saw.
Show Dark Factory observability dashboard — session metrics, quality scores, Ralph loop status, backlog summary.
Dark Factory — start/stop/status of the Ralph Wiggum autonomous loop, sync backlog, run single task, or validate sessions.
Initialize Dark Factory in a new project — creates .dark-factory/ directory, config.yaml, backlog, governance docs, and example holdout.
Evaluate project readiness for Dark Factory autonomous development. Scores 8 axes and produces a scorecard.
Run two-pass adversarial LLM quality evaluation on a completed feature implementation.
Executes bash commands
Hook triggers when Bash tool is used
Uses power tools
Uses Bash, Write, or Edit tools
Own this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge. GitHub access is read-only (username + org membership).
Sign in to claimOwn this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge. GitHub access is read-only (username + org membership).
Sign in to claimBased 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.
ship_modelShip strategy: direct-commit | pr-creation | pr-label (default: pr-creation)
${user_config.ship_model}github_repoGitHub repo in owner/repo format for issue sync and PR management
${user_config.github_repo}project_nameProject name for alerts and logs (e.g., XiriGo, Optiway)
${user_config.project_name}Marketplace for Claude Code plugins by Everva.
Open Claude Code (CLI, VS Code, or JetBrains) and run:
/plugin marketplace add everva/claude-plugins
This registers the Everva marketplace so you can install any plugin from it.
/plugin install dark-factory@everva # Autonomous AI pipeline
/plugin install mobile-agents@everva # iOS HIG + Android M3 specialists
/plugin install tdd-pipeline@everva # TDD tester + reviewer + UI polish
/plugin install code-review@everva # Code review with auto-fix
/reload-plugins
That's it. The plugin's skills, agents, and hooks are now available in your Claude Code session.
| Plugin | Description | Skills | Install |
|---|---|---|---|
| dark-factory | Autonomous AI development pipeline with governance tiers, holdout validation, satisfaction testing, and Ralph Wiggum loop | /dark-factory:factory, /dark-factory:task, /dark-factory:readiness, /dark-factory:dashboard, /dark-factory:init | /plugin install dark-factory@everva |
| mobile-agents | iOS and Android UI specialist agents — HIG compliance, Material Design 3, accessibility gates | Agents: ios-ui-specialist, android-ui-specialist | /plugin install mobile-agents@everva |
| tdd-pipeline | TDD pipeline agents — test-first tester, code reviewer, UI polish | Agents: tester, reviewer, ui-polish | /plugin install tdd-pipeline@everva |
| code-review | Multi-session code review with isolated reviewer sessions and auto-fix loop | /code-review:review, /code-review:review-fix | /plugin install code-review@everva |
Plugins are composable building blocks. Pick what your project needs:
Full-stack mobile project: Backend-only project: iOS-only app:
dark-factory dark-factory mobile-agents
mobile-agents tdd-pipeline tdd-pipeline
tdd-pipeline code-review code-review
+ project-specific agents + project-specific agents
Plugin agents are generic — they read your project's CLAUDE.md for context. Override any agent by placing a same-named .md file in your project's .claude/agents/ directory.
| Layer | Plugin | What it provides |
|---|---|---|
| Pipeline & Governance | dark-factory | Ralph loop, holdout validation, satisfaction judge, governance tiers, rate limiting, circuit breaker |
| Mobile Quality | mobile-agents | iOS HIG compliance, Android Material Design 3, accessibility gates |
| TDD & Review | tdd-pipeline | Test-first tester, code reviewer, UI polish agent |
| Code Review | code-review | Isolated review sessions, auto-fix loop |
# 1. Install
/plugin install dark-factory@everva
# 2. Initialize in your project (creates .dark-factory/ directory)
/dark-factory:init MyProject owner/repo
# 3. Edit risk factors for your project
# .dark-factory/config.yaml
# 4. Write holdout scenarios (agent-invisible behavioral tests)
# .dark-factory/holdouts/<layer>/*.holdout.yaml
# 5. Check project readiness score
/dark-factory:readiness
# 6. Run a single task through the full pipeline
/dark-factory:task docs/specs/my-feature.intent.md
# 7. Start the autonomous Ralph Wiggum loop
/dark-factory:factory start 10 8 # 10 tasks, 8 hours max
See Dark Factory README for full documentation.
# 1. Install
/plugin install code-review@everva
# 2. Review current changes
/code-review:review
# 3. Review + auto-fix loop (max 3 iterations)
/code-review:review-fix
See Code Review README for full documentation.
# 1. Install
/plugin install mobile-agents@everva
# 2. Agents are now available for spawning
# ios-ui-specialist — HIG compliance, VoiceOver, Dynamic Type
# android-ui-specialist — Material Design 3, TalkBack, Compose patterns
# 3. Override for your project (optional)
# Place a custom version in .claude/agents/ios-ui-specialist.md
# 1. Install
/plugin install tdd-pipeline@everva
# 2. Agents are now available for spawning
# tester — writes tests FIRST (TDD), coverage >= 80%
# reviewer — code quality gates, security checklist
# ui-polish — cross-platform consistency, micro-interactions
Add to ~/.claude/settings.json:
npx claudepluginhub everva/claude-plugins --plugin dark-factoryTDD pipeline agents: tester (test-first), reviewer (quality gates), and UI polish. Generic agents for test-driven development workflows.
Multi-session code review with auto-fix loop. Reviewer runs in isolated claude -p sessions for unbiased review, fixer stays in the current session to preserve project context.
Orchestrator skill — never does work itself, always dispatches Opus subagents (architect → tester → developer → reviewer). Two modes: interactive (at-keyboard) and away (overnight cron triage).
iOS and Android UI specialist agents for mobile development quality gates. HIG compliance for iOS, Material Design 3 for Android.
Feature development with code-architect/explorer/reviewer agents, CLAUDE.md audit and session learnings, and Agent Skills creation with eval benchmarking from Anthropic.
Production-grade engineering skills for AI coding agents — covering the full software development lifecycle from spec to ship.
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.
Complete developer toolkit for Claude Code
Intelligent draw.io diagramming plugin with AI-powered diagram generation, multi-platform embedding (GitHub, Confluence, Azure DevOps, Notion, Teams, Harness), conditional formatting, live data binding, and MCP server integration for programmatic diagram creation and management.
Orchestrate multi-agent teams for parallel code review, hypothesis-driven debugging, and coordinated feature development using Claude Code's Agent Teams