By cloudfieldcz
Enforce structured development workflows: transform vague ideas into technical analyses, generate implementation plans, follow test-driven development, systematically debug, and verify code with automated reviews and checks before merging.
MANDATORY before implementation. From idea to technical analysis — explores ideas through dialogue, then produces comprehensive analysis with cross-check reviews.
Execute plan in batches with review checkpoints
Create detailed implementation plan with bite-sized tasks
Use this agent when a technical analysis or specification needs review from a business analyst perspective. Examples: <example>Context: A technical analysis has been written. user: "The analysis for the email outbox feature is ready for review" assistant: "Let me dispatch the business-analyst-reviewer agent to check requirements completeness, user workflows, and edge cases" <commentary>Since a technical analysis document has been completed, use the business-analyst-reviewer agent to validate business requirements coverage.</commentary></example> <example>Context: A specification needs business validation before implementation planning. user: "Can you review the VAT rates analysis from a business perspective?" assistant: "I'll have the business-analyst-reviewer agent examine the analysis for completeness and business logic gaps" <commentary>The user explicitly requests business-perspective review of a technical document.</commentary></example>
Use this agent when a major project step has been completed and needs to be reviewed against the original plan and coding standards. Examples: <example>Context: The user is creating a code-review agent that should be called after a logical chunk of code is written. user: "I've finished implementing the user authentication system as outlined in step 3 of our plan" assistant: "Great work! Now let me use the code-reviewer agent to review the implementation against our plan and coding standards" <commentary>Since a major project step has been completed, use the code-reviewer agent to validate the work against the plan and identify any issues.</commentary></example> <example>Context: User has completed a significant feature implementation. user: "The API endpoints for the task management system are now complete - that covers step 2 from our architecture document" assistant: "Excellent! Let me have the code-reviewer agent examine this implementation to ensure it aligns with our plan and follows best practices" <commentary>A numbered step from the planning document has been completed, so the code-reviewer agent should review the work.</commentary></example>
Use this agent when a technical analysis or specification needs review from a developer perspective. Examples: <example>Context: A technical analysis has been written and needs technical validation. user: "The analysis for the VAT rates feature is ready -- please review the technical approach" assistant: "Let me dispatch the developer-reviewer agent to check technical feasibility, architecture, and integration points" <commentary>Since a technical analysis needs validation, use the developer-reviewer agent to verify technical soundness.</commentary></example> <example>Context: An analysis needs to be verified against actual code. user: "Can you verify the file references in the email outbox analysis are correct?" assistant: "I'll have the developer-reviewer agent read the codebase and verify all claims in the analysis" <commentary>The user wants code-level verification of the analysis -- the developer-reviewer reads actual files.</commentary></example>
Use this agent when a technical analysis, specification, or implementation needs review from a performance perspective. Examples: <example>Context: A technical analysis has been written. user: "Check the reporting feature analysis for performance issues" assistant: "Let me dispatch the performance-reviewer to check database patterns, indexes, and scalability" <commentary>Use the performance-reviewer for data-heavy features.</commentary></example> <example>Context: Code changes touch database queries or data processing. user: "Review the search implementation for performance" assistant: "I'll have the performance-reviewer check query patterns, indexes, and caching" <commentary>Search is performance-sensitive — check for N+1 queries and missing indexes.</commentary></example>
Use this agent when a technical analysis, specification, or implementation needs review from a security perspective. Examples: <example>Context: A technical analysis has been written. user: "Please review the API gateway analysis for security concerns" assistant: "Let me dispatch the security-reviewer agent to check trust boundaries, attack vectors, and authentication" <commentary>Use the security-reviewer to validate security posture.</commentary></example> <example>Context: Code changes touch authentication or data handling. user: "Review the session management changes for security" assistant: "I'll have the security-reviewer agent examine trust boundaries and bypass risks" <commentary>Security-critical paths need the security-reviewer.</commentary></example>
Use for non-trivial implementations involving design choices, multiple components, or unclear requirements. Explores ideas through dialogue, then produces a technical analysis with architecture, phases, risks, and cross-check reviews. SKIP for mechanical/single-file changes (Dockerfile, CI YAML, config tweak, one-liner) or when the user has already specified exactly what to build.
Use when facing 2+ independent tasks that can be worked on without shared state or sequential dependencies
Use when implementation of a feature, fix, or refactor is finished, before claiming work complete or starting merge/PR — any change that may affect docs/, README, CHANGELOG, or inline docstrings/JSDoc
Use when you have a written implementation plan to execute in a separate session with review checkpoints
Use when implementation is complete, all tests pass, and you need to decide how to integrate the work - guides completion of development work by presenting structured options for merge, PR, or cleanup
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.
No model invocation
Executes directly as bash, bypassing the AI model
No model invocation
Executes directly as bash, bypassing the AI model
Skills plugin for Claude Code that guides your AI coding agent through a structured development workflow — from idea to implementation.
Based on Superpowers by Jesse Vincent, extended with technical analysis workflow and cross-check review agents.
# 1. Register the marketplace
/plugin marketplace add cloudfieldcz/cf-powers
# 2. Install the plugin
/plugin install cf-powers@cf-powers
# 3. Verify — you should see cf-powers commands in the list
/help
That's it. The plugin activates automatically on every session.
Every feature starts the same way:
/analyse → /write-plan → execute
↓
idea → dialogue → analysis
↓
BA + Dev cross-check
/analyse — From idea to technical analysis in one step. Explores the idea through dialogue, then produces a full technical analysis (Czech output): architecture, DB changes, affected files, phases, risks, testing. Automatically dispatches BA and Developer reviewers for cross-check./write-plan — Break the analysis into bite-sized TDD implementation tasks.What happens at step 3 depends on the size of the feature:
When the analysis contains one logical phase, /write-plan produces a single plan file:
docs/plans/YYYY-MM-DD-feature-plan.md
You execute it in one go — either with subagent-driven development (in the current session) or by opening a new session with executing-plans (batch execution with checkpoints).
When the analysis defines multiple implementation phases, /write-plan produces one plan file per phase plus an index:
docs/plans/YYYY-MM-DD-feature-plan-index.md ← orchestration dashboard
docs/plans/YYYY-MM-DD-feature-plan-1-models.md ← phase 1
docs/plans/YYYY-MM-DD-feature-plan-2-services.md ← phase 2
docs/plans/YYYY-MM-DD-feature-plan-3-ui.md ← phase 3
You then execute phases one at a time (or in parallel if they have no dependencies). The index file tracks overall progress. After each phase completes, Claude updates the index and asks which phase to tackle next.
| Skill | When it activates |
|---|---|
| analysis | Non-trivial work with design choices — from idea to technical analysis |
| review-as-ba | Cross-check analysis from business analyst perspective |
| review-as-dev | Cross-check analysis from developer perspective |
| writing-plans | When you need a step-by-step implementation plan |
| executing-plans | Batch execution with human checkpoints |
| subagent-driven-development | Fast parallel execution with two-stage review |
| test-driven-development | During implementation (RED-GREEN-REFACTOR) |
| systematic-debugging | When encountering bugs or test failures |
| verification-before-completion | Before claiming work is done |
| requesting-code-review | After completing tasks |
| receiving-code-review | When processing review feedback |
| dispatching-parallel-agents | When facing independent parallel tasks |
| documenting-changes | After implementation, before merge/PR — keep docs in sync with code |
| finishing-a-development-branch | When ready to merge or create a PR |
| writing-skills | When creating new skills |
| using-superpowers | Injected automatically at session start |
/plugin update cf-powers
The SessionStart hook fail-closes if .claude-plugin/integrity.sha256
does not match the on-disk skill files. That means any release that
touches skills/*/SKILL.md must refresh the baseline before tagging,
otherwise installed clients will see a <security-alert> block instead
of the using-superpowers skill content after /plugin update.
Release checklist:
main.bin/update-integrity and commit
.claude-plugin/integrity.sha256 if it changed.version in both
.claude-plugin/plugin.json and
.claude-plugin/marketplace.json
following SemVer (skill additions/removals
= minor; behavior fixes = patch). Claude Code reads plugin.json
when deciding whether /plugin update has new content — if only
marketplace.json is bumped, clients silently skip the update.[Unreleased]
entries under the new version heading with today's date.tests/integrity/run-test.sh to confirm the hook still passes
on a clean tree.vX.Y.Z, push tag.If you forget step 2, end users get a fail-closed session on the next update. Recovery is to ship a follow-up release with the regenerated baseline.
npx claudepluginhub cloudfieldcz/cf-powers --plugin cf-powersSuperpowers Plus core skills library for Claude Code: planning, execution routing, TDD, debugging, and collaboration workflows
Harness-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
Core skills library for Claude Code: TDD, debugging, collaboration patterns, and proven techniques
Unity Development Toolkit - Expert agents for scripting/refactoring/optimization, script templates, and Agent Skills for Unity C# development
v9.44.1 — Patch release for Gemini environment/version detection and qwen auth gating. Run /octo:setup.
Claude harness - A harness for solo developers (Vibecoders) to handle full-cycle contract development.