By mikekelly
Deploy and manage Claude Code skills/agents to build software intelligently: implement via strict TDD, review/approve code, debug failures, run QA/tests, manage Docker environments, research online, design products, and analyze agent performance.
npx claudepluginhub mikekelly/promode --plugin promodeAnalyzes Claude Code agent behaviour from output files, primarily during after action reviews. Knows output format (JSON per line) and that tail -1 efficiently retrieves the final summary.
Reviews implementation work. Marks tasks done or requests rework. Use with model=sonnet for standard review, model=opus for complex architectural review.
Investigates failures, analyzes logs, and finds root causes. Produces a reproduction test and reports findings. Does NOT implement fixes unless explicitly asked. Use for debugging, logging analysis, and error investigation.
Manages dev environments: health checks, docker, services, scripts. Ensures environments are healthy and easily manageable.
Implements code using TDD workflow. Updates task docs with what was done. Commits changes before reporting.
Researches questions using web search, fetching, and MCPs (context7, grep_app, exa if available). Checks today's date first to ensure up-to-date results, especially for fast-moving areas like AI, developer tooling, and libraries.
Consults on user-facing product decisions. Thinks holistically: UX, psychology, behavioural economics, network effects, growth. Maintains design system in docs/product/.
Blackbox QA expert who runs through key scenarios from the outside in. Creates and maintains testing tools, vigilant about test performance.
Runs tests and returns AI-optimized results. Critiques test quality (naming, coverage, outside-in).
MUST be loaded before working with any Skill. Covers creating, building, reviewing, assessing, checking, auditing, evaluating, updating, modifying, and improving skills. Invoke PROACTIVELY before writing or changing any SKILL.md file. Provides structure, workflows, and validation for skill development. Supports both personal skills and standalone distributable skills (GitHub repos). (user)
MUST be loaded when setting up, installing, migrating, reviewing, auditing, or checking CLAUDE.md files in projects. Covers installing the promode CLAUDE.md into new projects, migrating existing CLAUDE.md content to AGENT_ORIENTATION.md (progressive disclosure), and auditing projects for conformance. Invoke PROACTIVELY when user mentions CLAUDE.md, project setup, agent configuration, or code meta files.
Install, update, list, and remove Claude Code skills. Supports GitHub repositories (user/repo), GitHub subdirectory URLs (github.com/user/repo/tree/branch/path), and .skill zip files. Use when user wants to install, add, download, update, sync, list, remove, uninstall, or delete skills.
Promode is a battle-tested methodology for AI-assisted software development. It covers the full lifecycle:
The methodology is designed for efficiency: conservative use of context, intelligent model selection for speed and cost, and aggressive delegation to keep the main conversation focused on high level objectives, not implementation detail.
Run two commands to get the plugin setup:
/plugin marketplace add mikekelly/promode
/plugin install promode
Then restart Claude Code and prompt Claude to install promode (update your CLAUDE.md, etc):
Update the claude code meta to install promode
The main agent handles high-level work: brainstorming with the user, designing features, planning implementation, and orchestrating execution. It delegates implementation to phase-specific agents:
| Agent | Purpose |
|---|---|
promode:implementer | TDD workflow, write code |
promode:reviewer | Code review, approve or request rework |
promode:tester | Run tests, parse results, critique quality |
promode:debugger | Root cause analysis, fix failures |
promode:environment-manager | Docker, services, health checks, scripts |
Each agent has the methodology embedded in its definition—they know TDD practices, behavioural-authority rules, and how to report results. Subagents commit changes before reporting back.
Tasks can run in parallel when independent, improving throughput on larger features.
Phase Agents — Implementation, review, and debugging agents with methodology baked in.
Skills:
MCP Servers (auto-start):
EXA_API_KEY)Promode makes skill management conversational:
Works with both user-level (~/.claude/skills/) and project-level (.claude/skills/) installations.
The development-workflow plugin for Claude Code — 35 skills organized around a 6-phase workflow (Think → Review → Build → Ship → Maintain → Setup), 24 agents, and an interactive setup wizard for rules, modes, hooks, and MCP servers.
Uses power tools
Uses Bash, Write, or Edit tools
Share bugs, ideas, or general feedback.
Comprehensive skills library for Claude Code: planning, design, TDD, debugging, collaboration patterns, and proven techniques
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.
Meta-prompting and spec-driven development system for Claude Code. Productivity framework for structured AI-assisted development.
Manus-style persistent markdown files for planning, progress tracking, and knowledge storage. Works with Claude Code, Kiro, Clawd CLI, Gemini CLI, Cursor, Continue, Hermes, and 17+ AI coding assistants. Now with Arabic, German, Spanish, and Chinese (Simplified & Traditional) support.
Orchestrate multi-agent teams for parallel code review, hypothesis-driven debugging, and coordinated feature development using Claude Code's Agent Teams