Add two new phases to improve plan quality through user collaboration:
- Phase 0: Idea Refinement - Refines feature ideas through dialogue
before running research agents. Asks questions one at a time,
prefers multiple choice, focuses on purpose/constraints/success
criteria. Skip option for detailed descriptions.
- Research Validation - After research completes, summarizes findings
and validates alignment before drafting the full plan.
Both phases are optional and can be skipped, ensuring the workflow
remains lightweight while catching misunderstandings early.
- Update "current year is 2025" to 2026 in 4 research agents
- Update year ranges to 2024-2026 for content searches in deepen-plan
and create-agent-skills workflows
- Change "currency" to "recency" for clarity in domain expertise skill
- Add branch detection at start of Setup Environment step
- Support continuing on existing feature branch or creating new
- Require explicit confirmation to commit to default branch
- Add incremental commit guidance with decision criteria table
- Include heuristic: "Can I write a meaningful commit message?"
- Generalize test commands to be framework-agnostic
- Remove quotes from the routing table's Response column
- Add blank line after <routing> tag to allow markdown parsing
GitHub treats custom XML tags as HTML, requiring a blank line
before markdown content to render correctly.
Co-authored-by: Claude Opus 4.5 <noreply@anthropic.com>
- Renamed every-marketplace to compound-marketplace
- Updated all EveryInc references to kieranklaassen
- Updated owner info in marketplace.json
Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
Added step to update original plan document with completed checkboxes
([ ] → [x]) after each task is finished. Keeps the plan as a living
document showing progress.
Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
- Updated PR template to include badge at bottom linking to plugin repo
- Added badge requirement to quality checklist
- Badge provides attribution and link to the plugin
Co-Authored-By: Claude <noreply@anthropic.com>
Skills are auto-loaded in context, no need to manually cat files.
Simplified Step 0 to just reference loaded skill context.
Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
- Added "Step 0: Discover and Load Design Skills (MANDATORY)"
- Discovers skills from all locations (plugin, global, project)
- Maps user context to relevant skills (Swiss → swiss-design, etc.)
- Changed iteration approach: ONE change per iteration, not 3-5
- Updated output format: focus on single most impactful change
- Added rule: "If you can't identify ONE clear improvement, stop"
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
- Add Phase 1: Discover and read relevant SKILL.md files before going online
- Add skill mappings for common technologies (Rails, Frontend, AI, etc.)
- Only proceed to online research if skills don't provide enough coverage
- Prioritize skill-based guidance with clear source attribution
- Synthesize findings from skills + online sources
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude <noreply@anthropic.com>
- Orchestrates: plan → deepen-plan → work → review → resolve todos → test-browser → feature-video
- Uses ralph-loop for autonomous completion
- Migrated from local command
- Updated to use /test-browser instead of /playwright-test
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude <noreply@anthropic.com>
- Add Step 0 asking user if they want to watch browser tests
- Headed mode opens visible browser window
- Headless mode runs faster in background
- Bump version to 2.25.0
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude <noreply@anthropic.com>
- Remove Playwright MCP server from plugin
- Add new agent-browser skill for CLI-based browser automation
- Rename /playwright-test to /test-browser command
- Update all commands and agents to use agent-browser CLI
- Update README and plugin.json
agent-browser is Vercel's headless browser CLI designed for AI agents.
It uses ref-based selection (@e1, @e2) from accessibility snapshots
and provides a simpler CLI interface compared to MCP tools.
Key benefits:
- No MCP server required
- Simpler Bash-based workflow
- Same ref-based element selection
- Better for quick automation tasks
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude <noreply@anthropic.com>
The /workflows:plan skill was producing inconsistent plan filenames
(sometimes gibberish, sometimes descriptive) because it lacked explicit
rules for converting issue titles to filenames.
Changes:
- Step 2: Added instruction to convert title to kebab-case filename
with example transformation (feat: Add User Auth → feat-add-user-auth.md)
- Output Format: Replaced vague <issue_title> placeholder with clear
pattern and good/bad examples
Co-authored-by: Claude Opus 4.5 <noreply@anthropic.com>
Enhanced /reproduce-bug command with browser automation:
- Added Phase 2 for visual bug reproduction using Playwright MCP
- Step-by-step guide for navigating to affected areas
- Screenshot capture at each reproduction step
- Console error checking
- User flow reproduction with clicks, typing, snapshots
- Better documentation structure with 4 clear phases
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude <noreply@anthropic.com>
All 26 agents now use `model: inherit` so they match whatever model
the user has configured in their main conversation.
Only `lint` keeps `model: haiku` since it's a simple task.
Fixes#69
Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
- New command for comprehensive agent-native architecture review
- Launches 8 parallel sub-agents, one per core principle
- Principles: Action Parity, Tools as Primitives, Context Injection,
Shared Workspace, CRUD Completeness, UI Integration, Capability
Discovery, Prompt-Native Features
- Each agent produces specific score (X/Y format with percentage)
- Generates summary report with overall score and top 10 recommendations
- Supports single principle audit via argument
Co-Authored-By: Claude <noreply@anthropic.com>
- Add low-quality preview GIF creation (640px, 128 colors)
- Use --s3-no-check-bucket flag for rclone uploads
- Document GitHub limitation: external MP4s don't embed
- Use clickable GIF linking to video for PR descriptions
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude <noreply@anthropic.com>
- Fix commands to use workflows: prefix
- Update install URL to compound-engineering-plugin
- Add philosophy section back
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
- Add MIT license (Copyright Every)
- Simplify README to focus on core workflow commands
- Link to full component reference and articles
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
The previous version incorrectly recommended XML tags for skill body content.
Anthropic's official specification uses standard markdown headings.
Changes:
- SKILL.md: Complete rewrite using markdown format, not XML tags
- Added references/official-spec.md from code.claude.com/docs/en/skills
- Added references/best-practices.md from platform.claude.com
- Removed obsolete use-xml-tags.md
- Updated naming to gerund form (creating-agent-skills)
- Descriptions now third person with "what" and "when"
BREAKING: If you followed the old XML tag guidance, convert to markdown headings.
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
- Read frontmatter (title, category, tags, module, symptom) of each learning
- Compare against plan to determine relevance
- SKIP learnings that are clearly not applicable
- SPAWN sub-agents only for learnings that MIGHT apply
- Example: 15 files found → filter to 3 relevant → spawn 3 sub-agents
Smarter than spawning 50 sub-agents when only 5 are relevant.
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
- Discover documented learnings from /workflows:compound
- Project docs/solutions/
- User ~/.claude/docs/
- Plugin docs/
- Spawn sub-agent for EACH learning to check if relevant to plan
- All learning sub-agents run in parallel
- Categories: performance-issues, debugging-patterns, config-fixes, etc.
- Added learnings to synthesis sources
- Fixed section numbering (now 1-9)
Institutional knowledge - prevents repeating past mistakes.
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
- Spawn a separate sub-agent for EVERY matched skill
- Each sub-agent reads and USES its assigned skill's SKILL.md
- All skill sub-agents run in parallel (10, 20, 30 is fine)
- Enhanced synthesis step to collect from ALL sources:
- Skill sub-agent outputs
- Research agents
- Review agents
- Context7 queries
- Web searches
- Deduplicate and prioritize findings before enhancing plan
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
- Dynamically discover ALL skills from all sources:
- Project .claude/skills/
- User ~/.claude/skills/
- compound-engineering plugin
- ALL other installed plugins
- Dynamically discover ALL agents from all sources:
- Project .claude/agents/
- User ~/.claude/agents/
- All installed plugins (not just compound-engineering)
- Local plugins
- Run ALL discovered agents in parallel (40+ is fine)
- No filtering by "relevance" - use everything available
- Match skills to plan sections and spawn sub-agents
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
- Add /deepen-plan command that enhances plans with parallel research agents
- Each plan section gets its own sub-agent for best practices, performance, UI research
- Update /workflows:plan to offer /deepen-plan as option 2 after plan creation
- Auto-run /deepen-plan when using ultrathink mode for maximum depth
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
* [2.17.0] Expand agent-native skill with mobile app learnings
Major expansion of agent-native-architecture skill based on real-world
learnings from building the Every Reader iOS app.
New reference documents:
- dynamic-context-injection.md: Runtime app state in system prompts
- action-parity-discipline.md: Ensuring agents can do what users can
- shared-workspace-architecture.md: Agents and users in same data space
- agent-native-testing.md: Testing patterns for agent-native apps
- mobile-patterns.md: Background execution, permissions, cost awareness
Updated references:
- architecture-patterns.md: Added Unified Agent Architecture, Agent-to-UI
Communication, and Model Tier Selection patterns
Enhanced agent-native-reviewer with comprehensive review process covering
all new patterns, including mobile-specific verification.
Key insight: "The agent should be able to do anything the user can do,
through tools that mirror UI capabilities, with full context about the
app state."
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude <noreply@anthropic.com>
* [2.18.0] Add Dynamic Capability Discovery and iCloud sync patterns
New patterns in agent-native-architecture skill:
- **Dynamic Capability Discovery** - For agent-native apps integrating with
external APIs (HealthKit, HomeKit, GraphQL), use a discovery tool (list_*)
plus a generic access tool instead of individual tools per endpoint.
(Note: Static mapping is fine for constrained agents with limited scope.)
- **CRUD Completeness** - Every entity needs create, read, update, AND delete.
- **iCloud File Storage** - Use iCloud Documents for shared workspace to get
free, automatic multi-device sync without building a sync layer.
- **Architecture Review Checklist** - Pushes reviewer findings earlier into
design phase. Covers tool design, action parity, UI integration, context.
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude <noreply@anthropic.com>
---------
Co-authored-by: Claude <noreply@anthropic.com>
Personalized coding tutorials that build on your existing knowledge
and use your actual codebase for examples. Includes spaced repetition
quizzes to reinforce learning.
- 3 commands: /teach-me, /quiz-me, /sync-tutorials
- 1 skill: coding-tutor
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-authored-by: Claude Opus 4.5 <noreply@anthropic.com>
Fixed issues flagged by skill-creator spec:
- Reference files now use proper markdown links instead of backtick text
- Descriptions use third person ("This skill should be used when...")
- Added Skill Compliance Checklist to CLAUDE.md with validation commands
Affected skills: agent-native-architecture, andrew-kane-gem-writer,
compound-docs, create-agent-skills, dhh-rails-style, dhh-ruby-style,
dspy-ruby, every-style-editor, file-todos, frontend-design, gemini-imagegen
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
- New `/xcode-test` command for building and testing iOS apps
- Uses XcodeBuildMCP to discover projects, build, install, launch
- Takes screenshots and captures console logs
- Supports human verification for Sign in with Apple, push, IAP
- Checks for XcodeBuildMCP installation before proceeding
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude <noreply@anthropic.com>
- New `/playwright-test` command for end-to-end browser tests on PR-affected pages
- Uses Playwright MCP to navigate, snapshot, check console errors
- Supports human-in-the-loop for OAuth/email/payment flows
- Creates P1 todos for failures and retries until passing
- Added Section 7 to `/workflows:review` - optional Playwright testing as subagent
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude <noreply@anthropic.com>
