Pirat83

gws-sheets-verify

Visual feedback loop for Google Sheets operations. After any write, take a Playwright screenshot and verify data integrity, sort order, and formatting.

Code & Development Listed

matplotlib-render-review

Render-Review-Refine workflow for matplotlib charts. Visual feedback loop using Claude Code's multimodal Read tool.

Rigorously triage a single Renovate or Dependabot dependency version-bump PR, in any ecosystem — Python/PyPI, Helm charts under Flux, container images, GitHub Actions, or npm. Pull the real changelog/release notes across the full old->new version range (not the registry's summary blurb), always read the actual source diff between the two tags as well — a changelog records what changed and can never record what stayed the same, which is what most "we are unaffected" arguments actually rest on — map both against actual usage sites in this repo, write a migration plan that distinguishes interface changes from conceptual/behavioral ones (a version bump can quietly change defaults or strictness with no signature change at all), verify whether existing checks would actually catch a regression in that exact spot or merely execute the line with stale data, write a regression test that fails on the old behavior when a real issue is found, fix what's fixable, report back linking the originating bot PR, and after the me

homelab-new-repo

Create a new Flux cluster-content repository in the Blue-Sharp home lab from the Blue-Sharp/homelab-template GitHub template — scaffolds the files, substitutes the NAME token, creates the namespace, mints the read-only Flux deploy key, re-seals the Discord webhook, and registers the repo with Flux. Use this whenever the user wants to add a new piece of software to the cluster, mentions "new repo", "new namespace", "add <software> to the cluster", "self-register with Flux", or asks how to onboard anything into the home lab — even if they don't name the template. Also use it when they ask what a new cluster repo needs, since the answer is "run this and it does the parts that can be automated". Do NOT use it for disaster recovery of EXISTING repos — that is flux-system/dr-reseed.sh.

stack-it

Manage a persistent stack of paused work — plans, progress, and loose ends — so it can be resumed in a later conversation. Use when the user says /stack-it, or when they want to pause current work to handle an interruption (bug fix, refactoring) and come back later. Supports push, pop, show, list, and drop subcommands. Also trigger when the user says things like "stack this", "save this for later", "park this work", or "resume stacked work".

AI & Automation Listed

tune-up

Concrete audit with structured output for the Claude Code workflow. Twelve peer-numbered phases: drift setup, permissions, agents, skills, plugins, MCP servers, hooks, rules, CLAUDE.md/coordinators/notification-sinks, summary, apply, post-apply quality gate. Each phase follows the same Subject/Sources/Checks/Output/Apply shape. Tracks per-project state in `.claude/.tune-up-state.json` so a project untouched for months produces a months-wide upstream-drift delta on its next run. Use this skill whenever the user wants to audit, clean up, optimise, or catch up on their Claude Code setup — even if they don't explicitly say `/tune-up`. Trigger phrases include "optimize my setup", "clean up Claude config", "audit my workflow", "what's new in Claude Code I should adopt", "my Claude config feels stale", "what plugins should I be using", and "drift check".