plan-utilities

Plugin plan-utilities from marketplace smirnovlabs-claude-skills in Smirnov-Labs/claude-skills. Commands read from the repo's manifest — yesterday.

Skills (22)

• adding-an-add-to-claude-button

  Use when adding an "Add to Claude" or "Connect to Claude" button or link for a remote MCP server, or when such a link opens an empty dialog, loses the server URL or points at //mcp.

• bd-to-br-migration

  Migrate docs from bd (beads) to br (beads_rust). Use when updating AGENTS.md, converting bd commands, "bd sync" → "br sync --flush-only", or beads migration.

• codex-plan-review

  Use when a written implementation plan needs an independent Codex review before execution: \"get a second opinion\", \"review plan with codex\", \"iterate on plan\".

• codex-review

  Use when you want an independent Codex review of a plan, spec or code change at a major checkpoint or before merging, when asked for a "codex review" or "second opinion", or before writing `codex exec` by hand.

• de-slopping

  Use before you finalize or call done a PR or commit message, design doc, README, runbook, report or release note, or when asked to de-slop, tighten or plain-English text.

• debugging-kubernetes-rollouts

  Use when a Kubernetes rollout or ArgoCD/GitOps sync fails: pods crashlooping or pending, image pull errors, a deployment stuck progressing, or a service that won't connect.

• debugging-nodejs-apps

  Use when a Node.js app won't start, hits npm or dependency errors, fails at runtime or can't connect, or when a simple-looking Node.js bug has survived several fixes.

• detect-overengineering

  Use when checking an implementation plan (docs/superpowers/plans/*.md or similar) for LLM-style overengineering before building it: \"check this plan for bloat\", \"is this overengineered\".

• evaluating-third-party-tools

  Use when someone asks whether an outside repo, tool, library, app or idea is useful to us, worth adopting or trying, or safe to install, before you download, install, build or run any of it.

• finding-latest-versions

  Use when you need the newest release of a GitHub repo or the newest tag or digest of a container image (Docker Hub, ghcr.io, quay.io, any OCI registry) before a bump or pin, instead of gh release list by hand.

• git-recon

  Use when the user wants a git-history health check of an unfamiliar or inherited repo: \"git recon\", \"codebase health check\", \"where are the landmines\", onboarding or due diligence. Main thread only.

• humanizer

  Remove signs of AI-generated writing from text. Use when editing or reviewing text to make it sound more natural and human-written. Based on Wikipedia's comprehensive "Signs of AI writing" guide. Detects and fixes patterns including: inflated symbolism, promotional language, superficial -ing analyse

• hyprland-lua

  Use when scripting or debugging Hyprland configured in Lua (hyprland.lua, 0.55+): Lua parse errors from `hyprctl dispatch` or `keyword`, a special workspace that toggles wrong or won't hide, workspace renames, a Quickshell bar not tracking workspaces.

• mining-claude-sessions

  Use when asked to find or resume a past Claude Code session ("find the session where we...", "resume the one that built X"), to say what a session did, or to audit Claude Code usage (which tools, skills, commands, agents or prompts recur), and before you parse ~/.claude/projects transcripts by hand.

• minto-pyramid

  Use when tightening or restructuring a draft (email, memo, status update, PR description, proposal) that opens with context, buries the ask, recaps in order or rambles, or when asked whether the lede is buried.

• nonstop

  Autonomous work mode that lets Claude work continuously while the user is away. Activates pre-flight risk assessment, dangerous-ops approval, and a blocker decision framework. Use "/nonstop" to activate, "/nonstop off" to deactivate. Use this skill whenever the user says things like "go nonstop", "k

• operating-frappe-on-kubernetes

  Use when operating, upgrading or debugging Frappe/ERPNext on Kubernetes: bench set-config that won't take, a first site's encryption_key, bench migrate, a major or Helm chart upgrade, Error Log, email or PDF errors.

• pr-fix

  Use when the user runs /pr-fix or asks to drive an open PR to clean by looping on CI failures and reviewer feedback (Codex, the Architecture Review / SAR, humans) — read, fix, commit, push, wait for the bots, repeat. It commits and pushes on its own, so it runs only on explicit request.

• pr-land

  Use when asked to merge or land a pull request ("merge it", "land it", "ship it", "merge when clean"), or to clean up the worktree and branch of a PR that is already merged.

• pr

  Use when the user runs /pr to commit the current work, push the branch, and open or update its pull request.

• presentation-blueprint

  Use when the user wants a presentation, pitch deck, slide deck or visual summary of a project, codebase, website or topic: \"build a deck\", \"make slides about\", \"pitch deck for\".

• watching-pr-bot-reviews

  Use when waiting for or reading a GitHub PR's bots (Codex, the Architecture Review or SAR, CI) with prbots: after a push, before merging, when asked \"what did the bots say\", or before writing a sleep loop around `gh api` or `gh pr view`.