By motsmanish
6-phase AI pipeline for shipping features into existing codebases. Multi-model consensus, hard-constraint phase gates, and verification that quotes file:line evidence before delivery.
Archive a completed motspilot task. Use when asked to archive a motspilot task or clean up finished work.
Initialize motspilot in the current project. Use when setting up motspilot for the first time in a new project directory.
Run the motspilot 6-phase AI pipeline for adding features to existing applications. Use when asked to run motspilot, run pipeline, or go motspilot.
Alias for /mots:pilot. Run the motspilot 6-phase AI pipeline. Use when asked to run pipeline.
Restore an archived motspilot task back to active status. Use when asked to reactivate, restore, or resume an archived motspilot task.
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.
GEMINI_API_KEYGoogle API key for consensus phase (Gemini model). Optional — consensus is skipped if missing.
${user_config.GEMINI_API_KEY}OPENAI_API_KEYOpenAI API key for consensus phase (GPT-4o model). Optional — consensus is skipped if missing.
${user_config.OPENAI_API_KEY}ANTHROPIC_API_KEYAnthropic API key — only required when CONSENSUS_CLAUDE_MODE=api. In the default session mode (inside Claude Code), Claude's consensus roles run via Task subagents and this key is not needed.
${user_config.ANTHROPIC_API_KEY}A 6-phase AI pipeline for shipping features into existing codebases — without the "the agent confidently broke prod" failure mode.
Each phase has a hard-constraint block the AI reads before any creative work, every verification finding quotes file:line evidence from your code, and a multi-model consensus (Claude + GPT-4o + Gemini) runs before the first design decision. Ships as a Claude Code plugin.
What you write: a 2–3 line feature description. What it produces: consensus synthesis → architecture plan with blast-radius analysis → code → tests (security-first) → verification report with confidence-scored findings → executed smoke tests → deployment plan. Each artifact is a file in your repo, so the full decision trail is committable.
See it before installing: docs/example-task/ walks through one feature end-to-end — every artifact from requirements through delivery, including the consensus synthesis, an architecture decomposition, a verification finding with file:line evidence, and executed smoke tests with side-effect checks.
motspilot is built for engineers integrating into production codebases — not for greenfield prototyping. If your codebase is younger than its first incident, a single-model agent is probably fine.
Two components:
motspilot.sh — Shell utility. Manages named tasks, requirements, state, and artifacts. Does not invoke AI directly.
Claude Code — The AI orchestrator. Reads the work order and runs each phase as a Task subagent. You approve each phase before it proceeds. Tasks auto-archive on completion.
Claude Code is the engine. motspilot.sh is the filing system.
brew install bashyq v4.52+ — parses YAML frontmatter from phase prompts
wget https://github.com/mikefarah/yq/releases/latest/download/yq_linux_amd64 -O ~/.local/bin/yq && chmod +x ~/.local/bin/yqbrew install yqFrom any Claude Code session:
/plugin marketplace add motsmanish/motspilot
/plugin install mots
Then in your target project:
/mots:init # One-time setup
/mots:pilot add login throttling # Run the pipeline
/mots:status # Check progress
/mots:view verify # Read verification report
/mots:archive --task=add-login-throttling # Archive when done
Requires: WSL, macOS, or Linux (bash). Consensus phase optionally requires PHP 8+ and API keys for Claude, GPT-4o, and Gemini.
git clone https://github.com/motsmanish/motspilot.git
See CONTRIBUTING.md for development setup. Alternative integration methods (symlink, submodule) are documented there.
# 1. Initialize (first time only)
./motspilot/motspilot.sh init
# Edit .motspilot/config with your project's language, framework, and test command
# 2. Create a task and prepare the pipeline
./motspilot/motspilot.sh go --task=csv-export "Add CSV export to the reports page"
# 3. In Claude Code chat:
run motspilot pipeline
Claude Code orchestrates all 6 phases (consensus → architecture → development → testing → verification → delivery) and archives the task automatically when delivery is approved. By default, phases run without pausing (AUTO_APPROVE="all"). Set AUTO_APPROVE="none" in .motspilot/config to approve each phase before it proceeds.
motspilot works with any framework. Framework-specific knowledge lives in prompts/frameworks/:
| Guide | Framework | Status |
|---|---|---|
cakephp.md | CakePHP 4.x | Shipped |
plain-php.md | Plain PHP (no framework) | Shipped |
laravel.md | Laravel 11.x+ | Shipped |
npx claudepluginhub motsmanish/motspilot --plugin motsComprehensive 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.
Develop, test, build, and deploy Godot 4.x games with Claude Code. Includes GdUnit4 testing, web/desktop exports, CI/CD pipelines, and deployment to Vercel/GitHub Pages/itch.io.
Comprehensive feature development workflow with specialized agents for codebase exploration, architecture design, and quality review
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.
Feature development with code-architect/explorer/reviewer agents, CLAUDE.md audit and session learnings, and Agent Skills creation with eval benchmarking from Anthropic.
Design fluency for frontend development. 1 skill with 23 commands (/impeccable polish, /impeccable audit, /impeccable critique, etc.) and curated anti-pattern detection.