All notable changes to the native-engineering plugin will be documented in this file.
The format is based on Keep a Changelog, and this project adheres to Semantic Versioning.
- Component Counts - Synchronized component counts across all metadata files:
- plugin.json: Updated to 29 agents, 25 commands, 32 skills
- marketplace.json: Updated to 29 agents, 25 commands, 32 skills
- README.md: Verified counts match actual files
- Skill Organization - Relocated
context-engineering-collectionskill to proper directory structure following project conventions (skills/context-engineering-collection/SKILL.md)
14 New Context Engineering Skills - Fully integrated from Agent-Skills-for-Context-Engineering as built-in skills (no external dependencies):
- Core Framework (
context-fundamentals,context-degradation,context-optimization,context-compression) - Architecture (
filesystem-context,memory-systems,multi-agent-patterns,tool-design,hosted-agents) - Development (
project-development,evaluation) - Advanced (
advanced-evaluation,bdi-mental-states) - Navigation (
context-engineering-collection- skill collection overview and learning paths)
Workflow Integration - Enhanced core workflows with context engineering execution triggers:
/workflow:work- Context health monitoring, position sensitivity guidance, compression triggers/workflow:review- Enhanced compound-recall-researcher with Context Clash detection/workflow:compound- Addedcontext-issues/category for documenting context-related problems/context-health- Built-in context degradation diagnostic command
- Removed External Dependencies - All context engineering capabilities are now built-in
- Updated Documentation -
docs/architecture/README.mdreflects built-in context engineering skills
- 28 agents, 25 commands, 29 skills (complete context-engineering framework integrated), 1 MCP server
- New Skills from Agent-Skills-for-Context-Engineering:
web-design-guidelines- Review UI code for Web Interface Guidelines compliance (fetches latest guidelines from Vercel)- Added
react-best-practicesto README documentation (45+ rules for React/Next.js performance optimization)
- 27 agents, 21 commands, 26 skills, 1 MCP server
- New Advanced Skills - Integrated 6 high-level skills from the Context Engineering Marketplace:
advanced-evaluation- Robust testing for agent reasoningbdi-mental-states- Intentional agent design (Belief, Desire, Intention)evaluation- Systematic benchmarking for LLM outputshosted-agents- Design for agents running in persistent environmentsmemory-systems- Advanced state management (Scratchpad/Summary/JSONL)tool-design- High-fidelity tool contract engineering
- Workflow: Review V2 - Minimalist, outcome-oriented code review workflow (
review-v2.md) - Session Tracking - Introduced
JSONLsession tracking pattern to documentation/workflows
- 27 agents, 21 commands, 24 skills, 1 MCP server
- add updating-plugin-version skill
- Localized Workflow Commands - Updated descriptions in command definition files (
commands/workflows/*.md) to Chinese. - Documentation - Updated component counts in README.
rcloneskill - Removed as it is not required for internal workflows.
- 27 agents, 20 commands, 13 skills, 1 MCP server
- Compound Recall System - Active knowledge recall from
docs/solutions/compound-recallskill - Dynamic indexing and retrieval of past solutions via Triple-Pass Search (Error -> Tag -> Keyword)compound-recall-researcheragent - Specialized researcher for institutional memory with template-driven reporting- Recall Templates - Standardized
recall-template.mdfor consistent knowledge injection - Automatic Indexing - Optimized
scripts/index_solutions.pywith keyword extraction and vector-based search - Optimization: Flash Recall Mechanism: Replaced static JSON indexing with dynamic on-demand scanning to eliminate Git conflicts in team environments.
- Improved Collaboration: Knowledge sedated in
docs/solutions/is now immediately discoverable by all team members aftergit pull. - Workflow Integration - Deeply integrated into
/workflow:plan,/workflow:work(error recovery),/workflow:review(pitfall checking), and/deepen-plan(memory lookup).
- 28 agents, 23 commands, 15 skills, 1 MCP server
/workflow:workcommand - PRs now include Compound Engineered badge- 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 that created the PR
design-iteratoragent - Now auto-loads design skills at start of iterations- Added "Step 0: Discover and Load Design Skills (MANDATORY)" section
- Discovers skills from ~/.claude/skills/, .claude/skills/, and plugin cache
- Maps user context to relevant skills (Swiss design → swiss-design skill, etc.)
- Reads SKILL.md files to load principles into context before iterating
- Extracts key principles: grid specs, typography rules, color philosophy, layout principles
- Skills are applied throughout ALL iterations for consistent design language
/test-browsercommand - Clarified to use agent-browser CLI exclusively- Added explicit "CRITICAL: Use agent-browser CLI Only" section
- Added warning: "DO NOT use Chrome MCP tools (mcp__claude-in-chrome__*)"
- Added Step 0: Verify agent-browser installation before testing
- Added full CLI reference section at bottom
- Added Next.js route mapping patterns
best-practices-researcheragent - Now checks skills before going online- Phase 1: Discovers and reads relevant SKILL.md files from plugin, global, and project directories
- Phase 2: Only goes online for additional best practices if skills don't provide enough coverage
- Phase 3: Synthesizes all findings with clear source attribution (skill-based > official docs > community)
- Skill mappings: Rails → dhh-rails-style, Frontend → frontend-design, AI → agent-native-architecture, etc.
- Prioritizes curated skill knowledge over external sources for trivial/common patterns
/lfgcommand - Full autonomous engineering workflow- Orchestrates complete feature development from plan to PR
- Runs: 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-browserinstead of/playwright-test
- 27 agents, 21 commands, 14 skills, 1 MCP server
agent-browserskill - Browser automation using Vercel's agent-browser CLI- Navigate, click, fill forms, take screenshots
- Uses ref-based element selection (simpler than Playwright)
- Works in headed or headless mode
-
Replaced Playwright MCP with agent-browser - Simpler browser automation across all browser-related features:
/test-browsercommand - Now uses agent-browser CLI with headed/headless mode option/feature-videocommand - Uses agent-browser for screenshotsdesign-iteratoragent - Browser automation via agent-browserdesign-implementation-revieweragent - Screenshot comparisonfigma-design-syncagent - Design verificationbug-reproduction-validatoragent - Bug reproduction/reviewworkflow - Screenshot capabilities/workworkflow - Browser testing
-
/test-browsercommand - Added "Step 0" to ask user if they want headed (visible) or headless browser mode
- Playwright MCP server - Replaced by agent-browser CLI (simpler, no MCP overhead)
/playwright-testcommand - Renamed to/test-browser
- 27 agents, 20 commands, 14 skills, 1 MCP server
/reproduce-bugcommand - Enhanced with Playwright visual reproduction:- Added Phase 2 for visual bug reproduction using browser automation
- Step-by-step guide for navigating to affected areas
- Screenshot capture at each reproduction step
- Console error checking
- User flow reproduction with clicks, typing, and snapshots
- Better documentation structure with 4 clear phases
- 27 agents, 21 commands, 13 skills, 2 MCP servers
- Agent model inheritance - All 26 agents now use
model: inheritso they match the user's configured model. Onlylintkeepsmodel: haikufor cost efficiency. (fixes #69)
- 27 agents, 21 commands, 13 skills, 2 MCP servers
/agent-native-auditcommand - 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
- 27 agents, 21 commands, 13 skills, 2 MCP servers
rcloneskill - Upload files to S3, Cloudflare R2, Backblaze B2, and other cloud storage providers
/feature-videocommand - Enhanced with:- Better ffmpeg commands for video/GIF creation (proper scaling, framerate control)
- rclone integration for cloud uploads
- Screenshot copying to project folder
- Improved upload options workflow
- 27 agents, 20 commands, 13 skills, 2 MCP servers
- Version history cleanup after merge conflict resolution
This release consolidates all recent work:
/feature-videocommand for recording PR demos/deepen-plancommand for enhanced planningcreate-agent-skillsskill rewrite (official spec compliance)agent-native-architectureskill major expansiondhh-rails-styleskill consolidation (merged dhh-ruby-style)- 27 agents, 20 commands, 12 skills, 2 MCP servers
/feature-videocommand - Record video walkthroughs of features using Playwright
create-agent-skillsskill - Complete rewrite to match Anthropic's official skill specification
dhh-ruby-styleskill - Merged intodhh-rails-styleskill
-
/deepen-plancommand - Power enhancement for plans. Takes an existing plan and runs parallel research sub-agents for each major section to add:- Best practices and industry patterns
- Performance optimizations
- UI/UX improvements (if applicable)
- Quality enhancements and edge cases
- Real-world implementation examples
The result is a deeply grounded, production-ready plan with concrete implementation details.
/workflow:plancommand - Added/deepen-planas option 2 in post-generation menu. Added note: if running with ultrathink enabled, automatically run deepen-plan for maximum depth.
-
agent-native-architectureskill - Added Dynamic Capability Discovery pattern and Architecture Review Checklist:New Patterns in mcp-tool-design.md:
- Dynamic Capability Discovery - For external APIs (HealthKit, HomeKit, GraphQL), build a discovery tool (
list_*) that returns available capabilities at runtime, plus a generic access tool that takes strings (not enums). The API validates, not your code. This means agents can use new API capabilities without code changes. - CRUD Completeness - Every entity the agent can create must also be readable, updatable, and deletable. Incomplete CRUD = broken action parity.
New in SKILL.md:
- Architecture Review Checklist - Pushes reviewer findings earlier into the design phase. Covers tool design (dynamic vs static, CRUD completeness), action parity (capability map, edit/delete), UI integration (agent → UI communication), and context injection.
- Option 11: API Integration - New intake option for connecting to external APIs like HealthKit, HomeKit, GraphQL
- New anti-patterns: Static Tool Mapping (building individual tools for each API endpoint), Incomplete CRUD (create-only tools)
- Tool Design Criteria section added to success criteria checklist
New in shared-workspace-architecture.md:
- iCloud File Storage for Multi-Device Sync - Use iCloud Documents for your shared workspace to get free, automatic multi-device sync without building a sync layer. Includes implementation pattern, conflict handling, entitlements, and when NOT to use it.
- Dynamic Capability Discovery - For external APIs (HealthKit, HomeKit, GraphQL), build a discovery tool (
This update codifies a key insight for agent-native apps: when integrating with external APIs where the agent should have the same access as the user, use Dynamic Capability Discovery instead of static tool mapping. Instead of building read_steps, read_heart_rate, read_sleep... build list_health_types + read_health_data(dataType: string). The agent discovers what's available, the API validates the type.
Note: This pattern is specifically for agent-native apps following the "whatever the user can do, the agent can do" philosophy. For constrained agents with intentionally limited capabilities, static tool mapping may be appropriate.
-
agent-native-architectureskill - Major expansion based on real-world learnings from building the Every Reader iOS app. Added 5 new reference documents and expanded existing ones:New References:
- dynamic-context-injection.md - How to inject runtime app state into agent system prompts. Covers context injection patterns, what context to inject (resources, activity, capabilities, vocabulary), implementation patterns for Swift/iOS and TypeScript, and context freshness.
- action-parity-discipline.md - Workflow for ensuring agents can do everything users can do. Includes capability mapping templates, parity audit process, PR checklists, tool design for parity, and context parity guidelines.
- shared-workspace-architecture.md - Patterns for agents and users working in the same data space. Covers directory structure, file tools, UI integration (file watching, shared stores), agent-user collaboration patterns, and security considerations.
- agent-native-testing.md - Testing patterns for agent-native apps. Includes "Can Agent Do It?" tests, the Surprise Test, automated parity testing, integration testing, and CI/CD integration.
- mobile-patterns.md - Mobile-specific patterns for iOS/Android. Covers background execution (checkpoint/resume), permission handling, cost-aware design (model tiers, token budgets, network awareness), offline handling, and battery awareness.
Updated References:
- architecture-patterns.md - Added 3 new patterns: Unified Agent Architecture (one orchestrator, many agent types), Agent-to-UI Communication (shared data store, file watching, event bus), and Model Tier Selection (fast/balanced/powerful).
Updated Skill Root:
- SKILL.md - Expanded intake menu (now 10 options including context injection, action parity, shared workspace, testing, mobile patterns). Added 5 new agent-native anti-patterns (Context Starvation, Orphan Features, Sandbox Isolation, Silent Actions, Capability Hiding). Expanded success criteria with agent-native and mobile-specific checklists.
-
agent-native-revieweragent - Significantly enhanced with comprehensive review process covering all new patterns. Now checks for action parity, context parity, shared workspace, tool design (primitives vs workflows), dynamic context injection, and mobile-specific concerns. Includes detailed anti-patterns, output format template, quick checks ("Write to Location" test, Surprise test), and mobile-specific verification.
These updates operationalize a key insight from building agent-native mobile apps: "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." The failure case that prompted these changes: an agent asked "what reading feed?" when a user said "write something in my reading feed"—because it had no publish_to_feed tool and no context about what "feed" meant.
dhh-rails-styleskill - Massively expanded reference documentation incorporating patterns from Marc Köhlbrugge's Unofficial 37signals Coding Style Guide:- controllers.md - Added authorization patterns, rate limiting, Sec-Fetch-Site CSRF protection, request context concerns
- models.md - Added validation philosophy, let it crash philosophy (bang methods), default values with lambdas, Rails 7.1+ patterns (normalizes, delegated types, store accessor), concern guidelines with touch chains
- frontend.md - Added Turbo morphing best practices, Turbo frames patterns, 6 new Stimulus controllers (auto-submit, dialog, local-time, etc.), Stimulus best practices, view helpers, caching with personalization, broadcasting patterns
- architecture.md - Added path-based multi-tenancy, database patterns (UUIDs, state as records, hard deletes, counter caches), background job patterns (transaction safety, error handling, batch processing), email patterns, security patterns (XSS, SSRF, CSP), Active Storage patterns
- gems.md - Added expanded what-they-avoid section (service objects, form objects, decorators, CSS preprocessors, React/Vue), testing philosophy with Minitest/fixtures patterns
- Reference patterns derived from Marc Köhlbrugge's Unofficial 37signals Coding Style Guide
- All skills - Fixed spec compliance issues across 12 skills:
- Reference files now use proper markdown links (
[file.md](./references/file.md)) instead of backtick text - Descriptions now use third person ("This skill should be used when...") per skill-creator spec
- Affected skills: agent-native-architecture, andrew-kane-gem-writer, compound-docs, create-agent-skills, dhh-rails-style, dspy-ruby, every-style-editor, file-todos, frontend-design, gemini-imagegen
- Reference files now use proper markdown links (
- CLAUDE.md - Added Skill Compliance Checklist with validation commands for ensuring new skills meet spec requirements
/workflow:reviewcommand - Section 7 now detects project type (Web, iOS, or Hybrid) and offers appropriate testing. Web projects get/playwright-test, iOS projects get/xcode-test, hybrid projects can run both.
-
/xcode-testcommand - Build and test iOS apps on simulator using XcodeBuildMCP. Automatically detects Xcode project, builds app, launches simulator, and runs test suite. Includes retries for flaky tests. -
/playwright-testcommand - Run Playwright browser tests on pages affected by current PR or branch. Detects changed files, maps to affected routes, generates/runs targeted tests, and reports results with screenshots.