home-manager

Coordinate full-strength Claude and Codex teammates in detached tmux sessions with external stall monitoring, durable file channels, completion markers, and resumable conversations. Use when work can be divided into independent tasks, when the user asks for teammates or subagents, or when delegating research, implementation, frontend, review, or long-running work across model accounts.

Coordinate full-strength Claude and Codex teammates in detached tmux sessions with external stall monitoring, durable file channels, completion markers, and resumable conversations. Use when work can be divided into independent tasks, when the user asks for teammates or subagents, or when delegating research, implementation, frontend, review, or long-running work across model accounts.

Drive the kautopilot binary as a thin in-session controller: ticket/free-form request → merge-ready PRs across one or more repos, as ONE flat session. The binary owns all state, prompts, and next-step decisions; this skill just runs what `kautopilot next` yields. Use when running /kautopilot.

Drive the kautopilot binary as a thin in-session controller: ticket/free-form request → merge-ready PRs across one or more repos, as ONE flat session. The binary owns all state, prompts, and next-step decisions; this skill just runs what `kautopilot next` yields. Use when running /kautopilot.

Iteratively review-and-resolve the current changes using parallel subagent reviewers across multiple lenses (bugs, quality, completion), looping until every reviewer is satisfied. Use when running /floop, or when asked to deeply review and harden a diff until all critiques are addressed.

Give the user a fast, ADHD/dyslexia-friendly recap of the current work or conversation — the outcome first, then what (if anything) you need from them. Use when running /summary, or when asked to "summarize", "recap", "where are we", "catch me up", or "what do you need from me".

Iteratively review-and-resolve the current changes using parallel subagent reviewers across multiple lenses (bugs, quality, completion), looping until every reviewer is satisfied. Use when running /floop, or when asked to deeply review and harden a diff until all critiques are addressed.

Give the user a fast, ADHD/dyslexia-friendly recap of the current work or conversation — the outcome first, then what (if anything) you need from them. Use when running /summary, or when asked to "summarize", "recap", "where are we", "catch me up", or "what do you need from me".

Open a pull request for the current branch and babysit it to ready-to-merge: push, create/reuse the PR, then loop — fix failing CI, keep the branch current with base, and triage + answer every review thread (CodeRabbit and human) until CI is green and all threads are resolved. Never merges. Use when running /babysit-pr, or any "create the PR and watch it green up" request, standalone (no kautopilot session needed).

Create tasks in ClickUp with support for status changes, dependencies, assignees, and tags. Use when the user asks to create a task, change task status, add a dependency, or set up task relationships in ClickUp.

Open a pull request for the current branch and babysit it to ready-to-merge: push, create/reuse the PR, then loop — fix failing CI, keep the branch current with base, and triage + answer every review thread (CodeRabbit and human) until CI is green and all threads are resolved. Never merges. Use when running /babysit-pr, or any "create the PR and watch it green up" request, standalone (no kautopilot session needed).

Create tasks in ClickUp with support for status changes, dependencies, assignees, and tags. Use when the user asks to create a task, change task status, add a dependency, or set up task relationships in ClickUp.

Create a new agent skill in the open SKILL.md standard (works with Claude Code AND Codex). Use when asked to create a skill, make a skill, or build a reusable agent workflow/automation. Generates a SKILL.md plus optional supporting files.

Create distinctive, production-grade frontend interfaces with high design quality. Use this skill when the user asks to build web components, pages, posters, or applications (examples include websites, landing pages, dashboards, React components, HTML/CSS layouts, or when styling/beautifying any web UI). Generates creative, polished code and UI design that avoids generic AI aesthetics.

ArgoCD via loctl. Use when checking deployment status, sync health, or app history.

AWS and Terraform via loctl. Use when checking AWS resources, S3 buckets, EKS clusters, or running terraform plans.

Kubernetes cluster access via loctl (kubectl, helm, stern). Use when checking pod health, inspecting resources, or streaming logs from pods.

Observability via loctl (promtool, grafanactl, logs). Use when querying metrics in Mimir, logs in Loki, or browsing Grafana dashboards.

Umbrella entry point for Liftoff infrastructure access via loctl: Kubernetes, Mimir, Loki, Grafana, ArgoCD, AWS, Helm, Terraform. Use when investigating PE tickets or debugging production infrastructure, then load the matching liftoff-* sub-skill for detail. Never call kubectl/helm/aws/argocd directly.

Start or hand off a named assistant session inside zellij. Use when the user runs /rc-session, says "start a session" or "start an rc session", or wants to switch between mobile/browser SSH and CLI for Claude remote-control or Codex. Use `klaude handoff` for Claude and `kodex handoff` for Codex. Do not create or ask about worktrees.

Create a new agent skill in the open SKILL.md standard (works with Claude Code AND Codex). Use when asked to create a skill, make a skill, or build a reusable agent workflow/automation. Generates a SKILL.md plus optional supporting files.

Create distinctive, production-grade frontend interfaces with high design quality. Use this skill when the user asks to build web components, pages, artifacts, posters, or applications (examples include websites, landing pages, dashboards, React components, HTML/CSS layouts, or when styling/beautifying any web UI). Generates creative, polished code and UI design that avoids generic AI aesthetics.

Create a new agent skill in the open SKILL.md standard (works with Claude Code AND Codex). Use when asked to create a skill, make a skill, or build a reusable agent workflow/automation. Generates a SKILL.md plus optional supporting files.

Drive the kautopilot binary as a thin in-session controller: ticket/free-form request → merge-ready PRs across one or more repos, as ONE flat session. The binary owns all state, prompts, detection, and next-step decisions; this skill just runs what `kautopilot next` yields. Use when running /kautopilot.

Iteratively review-and-resolve the current changes using parallel subagent reviewers across multiple lenses (bugs, quality, completion), looping until every reviewer is satisfied. Use when running /floop, or when asked to deeply review and harden a diff until all critiques are addressed.

Start a session: spin up a detached Claude Code remote-control (--rc) session via a yolo-* wrapper, in a worktrunk worktree, inside zellij. Use when the user says "start a session" / "start a new session", runs /rc-session, wants a remote-control / yolo session to drive remotely, or is kicking off work on a ticket in a fresh worktree.

Open a pull request for the current branch and babysit it to ready-to-merge: push, create/reuse the PR, then loop — fix failing CI, keep the branch current with base, and triage + answer every review thread (CodeRabbit and human) until CI is green and all threads are resolved. Never merges. Use when running /babysit-pr, or any "create the PR and watch it green up" request, standalone (no kautopilot session needed).

Use loctl when accessing infrastructure: Kubernetes, Mimir, Loki, Grafana, ArgoCD, Vault, AWS, Helm, Terraform. Use when investigating PE tickets, checking pod health, querying metrics/logs, or debugging production issues.

Create tasks in ClickUp with support for status changes, dependencies, assignees, and tags. Use when the user asks to create a task, change task status, add a dependency, or set up task relationships in ClickUp.

ArgoCD via loctl. Use when checking deployment status, sync health, or app history.

AWS and Terraform via loctl. Use when checking AWS resources, S3 buckets, EKS clusters, or running terraform plans.

Kubernetes cluster access via loctl (kubectl, helm, stern). Use when checking pod health, inspecting resources, or streaming logs from pods.

Observability via loctl (promtool, grafanactl, logs). Use when querying metrics in Mimir, logs in Loki, or browsing Grafana dashboards.

Find and remove dead code using Knip (expansive) + TS Language Server verification. Use when running /kill-deadcode, removing unused exports/types, or cleaning up dead code.