eval-loop
01Command Claude Code
You are running an evaluator-optimizer loop on the code that was just written or modified in this session. This implements a structured generate-evaluate-refine cycle with a maximum of 3 iterations.
Command Claude Code
You are running an evaluator-optimizer loop on the code that was just written or modified in this session. This implements a structured generate-evaluate-refine cycle with a maximum of 3 iterations.
Hook Claude Code
Runs when a session starts, executing session-register.sh. From abuango/pos-ai.
Hook Claude Code
Runs when a session ends, executing session-close.sh. From abuango/pos-ai.
Hook Claude Code
Runs when you submit a prompt, before the agent sees it, executing skill-eval.sh. From abuango/pos-ai.
Hook Claude Code
Runs before the agent uses a tool for Edit and Write tool calls, executing plan-gate.sh with --show-current (2 commands). From abuango/pos-ai.
Hook Claude Code
Runs after a tool call finishes for Edit and Edit tool calls, executing .test.ts with --write, --noEmit, --testPathPattern and --passWithNoTests (2 commands). From abuango/pos-ai.
Hook Claude Code
Runs when a task is marked completed, running an inline shell check. From abuango/pos-ai.
Settings file Claude Code
POS workflow settings for Claude Code.
Skill Claude CodeCodex
Review API design for consistency, naming conventions, versioning, and best practices. Use when designing APIs, reviewing endpoints, or when user mentions "API review", "API design", "endpoint review", "REST API", "API conventions", or "API consistency".
Skill Claude CodeCodex
System design review, architecture decision records, and component structure analysis. Use when designing systems, reviewing architecture, making technology decisions, or when user mentions "architecture", "system design", "ADR", "component structure", or "design review".
Skill Claude CodeCodex
Perform standardized code reviews for projects. Use when reviewing code, checking implementation quality, or when user mentions "code review", "review changes", "check the code", or "review implementation".
Skill Claude CodeCodex
Draft professional communications. Use when writing emails, Slack messages, meeting follow-ups, status updates, or any professional correspondence. Triggers on "draft email", "write message", "meeting follow-up", "send update", or "compose response".
Skill Claude CodeCodex
Review written content (blog posts, docs, articles) for AI-sounding language and apply technical writing best practices. Use when writing or reviewing blog posts, articles, documentation, or any written content. Triggers on "review content", "check writing", "content review", "does this sound like AI", "writing…
Skill Claude CodeCodex
Create standardized pull/merge requests with summary, test plan, and linked context. Use when creating a PR, opening a merge request, or when user mentions "create PR", "open MR", "merge request", "pull request", "submit for review", or "push and create PR".
Skill Claude CodeCodex
Create and manage cross-agent handoff records. Use when handing off work to another agent/model, creating task records, or when user mentions 'handoff', 'hand off', 'pass to another agent', 'create handoff'.
Skill Claude CodeCodex
Learning Coach skill for creating structured learning curricula. Use when designing study plans, certification prep paths, or skill development programs. Triggers on phrases like "create curriculum", "study plan for", "learn X", "certification prep", or "design learning path".
Skill Claude CodeCodex
Systematic debugging and troubleshooting playbook. Use when diagnosing bugs, investigating errors, tracing issues, or when user mentions "debug", "troubleshoot", "investigate", "error", "bug", "broken", "not working", or "trace issue".
Skill Claude CodeCodex
Generate Excalidraw architecture diagrams from natural language descriptions. Use when creating system diagrams, architecture visualizations, data flow diagrams, or when user mentions "diagram", "excalidraw", "draw architecture", "visualize system", or "generate diagram".
Skill Claude CodeCodex
Generate persistent feature specification documents that serve as the source of truth across sessions. Use when specifying features, documenting requirements, or when user mentions "feature spec", "write spec", "specification", "requirements doc", "define feature", or "spec this".
Skill Claude CodeCodex
Create high-quality, distinctive web interfaces that avoid generic AI aesthetics. Use when building web components, pages, dashboards, landing pages, or applications. Triggers on "build UI", "create page", "design frontend", "make it look good", "dashboard design", or any web interface creation task.
Skill Claude CodeCodex
Access CTO system documentation and learn how components work. Use when user asks 'how does X work', 'show me how to', 'explain the', 'what is', 'learn about', '/learn', or needs help understanding the system.
Skill Claude CodeCodex
Run code-review, security-audit, and documentation review in parallel using agents. Use when reviewing completed work, pre-merge checks, or when user mentions 'parallel review', 'full review', 'review everything', or 'comprehensive review'.
Skill Claude CodeCodex
Generate implementation plans for projects. Use when creating a plan, drafting implementation steps, or when user mentions "create plan", "implementation plan", "planning phase", or "plan this project".
Skill Claude CodeCodex
Execute product marketing workflows including GTM strategy, positioning, competitive analysis, campaign planning, and launch coordination. Use when creating marketing plans, positioning documents, GTM strategies, or when user mentions "product marketing", "GTM", "go to market", "positioning", "competitive analysis"…