lexfrei/ccc

End-to-end closeout of unresolved review comments across one or more GitHub PRs — verifies each thread, fixes legitimate findings with one signed commit per thread, posts approved replies, restores original branch

Interactive skill for creating well-structured Claude Code subagents following official documentation standards

Where's the proof, Billy? — stop and prove a claim with deep investigation

Helm chart TDD specialist — creates production-ready Helm charts with helm-unittest following .architecture.yaml standards

Local disk cleanup utilities: stale git worktrees, dead personal forks, Docker prune (incl. non-default buildx builders and VM disk trim), Go caches, and Homebrew — each runs report → confirm → execute with interactive prompts, plus one orchestrating pipeline

Code quality validation agent — runs linters, tests, security checks, and verifies .architecture.yaml compliance

AI artifact cleanup agent — detects and removes AI-generated comments, excessive documentation, and non-professional code patterns

Containerization specialist — creates optimized, secure Containerfiles with multi-stage builds and distroless base images

Generate a /rename command from type, title, and optional PR number — pure string formatting utility

Git workflow utilities: fork refresh and repo setup with branch protection

Auto-run golangci-lint on package when Go files are edited

Go developer agent — TDD specialist for cloud-native Go applications with Echo, slog, and cockroachdb/errors

Add or modernize Gateway API *Route templates (HTTPRoute/GRPCRoute/TLSRoute/TCPRoute/UDPRoute) in a Helm chart, mirroring its ingress conventions. Auto-detects add vs update mode per route kind, applies current best practices (named rules, typed filters, optional BackendTLSPolicy). File edits only — no git operations, no push, no PR

Kubernetes specialist — creates secure, production-ready K8s manifests and ArgoCD applications with zero-trust networking

Analyze the current session and extract key learnings into CLAUDE.md files and memory system

Assemble m4b audiobook from audio files with chapters, metadata, and cover art using ffmpeg

Session status hooks for the macro-claude macropad plugin — writes per-session JSON so the MX Creative Console shows live Claude Code state

MCP server for querying Grafana Loki logs — LogQL queries, label discovery, series exploration, and index statistics

MCP server for Telegram Client API (MTProto) — messaging, contacts, groups, channels, media, stickers via user account

MCP server for managing Transmission BitTorrent client — list, add, remove, start, stop torrents, session stats, queue and bandwidth management

PreToolUse hook: reject Write/Edit on markdown files when a paragraph spans more than one line. Enforces 'one continuous line per paragraph' so renderers wrap to viewer width instead of fighting hardwraps baked into the source

Measure porosity, pore size/shape, lattice period and ordering (psi6, Voronoi coordination, radial g(r) and orientational correlation length) from top-view SEM images of porous films such as anodic aluminium oxide. The agent reads the image to calibrate and crop; a Python core does the morphometry and flags results that contradict physics. Vision decides, Python computes — no hardcoded autodetect, so it adapts to different instruments and scale-bar formats.

Run Renovate locally in dry-run mode to discover available dependency updates, research migration guides for major bumps, and apply selected updates

Comprehensive PR review pipeline: a pre-review readiness gate (stops on merge conflicts or red code-related CI with a fix-it note, no verdict), branch-level analysis with merge-base attribution (the verdict is decided only by defects the diff introduced, worsened, or interacts with — purely pre-existing defects are reported in full in a separate outside-findings section and routed to a filed issue, parked branch, or memory entry instead of blocking), sequential Claude+Codex dual-model review with evidence-based verification, and opt-in GitHub PR review publishing under a five-frame substance framework. A value/design gate weighs whether the change should exist at all — value versus permanent maintenance cost, right layer, root cause in-repo versus upstream — and can block a PR even when the code is flawless. Remedies stay proportional: changed logic needs tests in the existing test surface, fixed prose is its own deliverable, and a new checker or guard script is never a required remedy. Findings carry git-computed origin labels (feature vs churn) that diagnose diverging review loops without ever touching the verdict. Reviews land on an LGTM or NOT LGTM verdict, compatible with non-blocking recommendations; the only no-verdict path is the readiness gate's fix-CI / fix-conflicts note.

Speak text aloud using macOS TTS with automatic voice selection

Copy markdown to macOS clipboard as rich HTML + plain-text fallback, ready for Cmd+V into Slack with formatting preserved

Python developer agent — TDD specialist treating code as art, with FastAPI, Pydantic, structlog, and strict typing

Project analysis and planning agent — decomposes complex tasks, creates implementation plans, identifies risks and dependencies

Technical architect agent — manages .architecture.yaml, creates ADRs, evaluates technology choices and design patterns

Frontend specialist — Go Templ + HTMX with server-first rendering and WCAG 2.1 AA accessibility

Generate concise plain-text TLDR summaries for GitHub Pull Requests, ready for pasting into Slack or other messengers

Automatic daily work-sync log — PostToolUse hook captures gh pr/issue create/merge/close, and a companion skill lets Claude log research/ops/checkpoint events into a dated markdown note (feeds standups via Obsidian/Logseq/plain markdown)