claude-code-skills

Use when asked to document work, record decisions, create session records, write ADRs or RFCs, or update knowledge notes in ~/knowledge. Also use when confused about who a person is, what a company is, what situation the user is referring to, or any personal or professional context — read the vault before asking.

Use when creating or editing notes in an Obsidian vault, writing to ~/knowledge, or using Obsidian-specific syntax (wikilinks, callouts, embeds, frontmatter properties).

Use when reviewing any implementation. Replaces fixed spec+quality two-step with category-driven parallel specialists: an assessor picks relevant categories for the specific artifact, one specialist per category reviews in parallel, controller aggregates. Use after implementer completes a task in subagent-driven-development.

Use when dispatching work to background agents, defining custom subagents, or choosing between local agent dispatch and VM-based workers (claude-ctl).

Use when working with Kubernetes clusters, GitOps deployments, Flux reconciliation, Helm releases, or cluster troubleshooting. Guides declarative cluster management and GitOps workflows. Not for secrets encryption or SOPS operations — route those to secrets-management.

Use when sending a notification to ntfy from a homelab service, loop, or agent. Covers the canonical in-cluster publish pattern and the silent-failure trap of posting to the Authentik-gated public URL.

Use when encrypting secrets with SOPS/age, managing Kubernetes Secret manifests, rotating credentials, or setting up secret delivery (sops-nix, sealed secrets). Not for application code that reads env vars or auth tokens at runtime.

Use when writing code, reviewing PRs, refactoring, or making architectural decisions. Enforces readability, testability, and maintainability standards across all languages.

Use when building container images, writing Dockerfiles, pushing to registries, or optimizing image size and security. Guides rootless builds, multi-stage patterns, and supply chain security.

Use when working with NixOS configurations, Nix flakes, module composition, system rebuilds, or Home Manager. Guides declarative system management patterns and safe rebuild workflows.

Use when adding metrics, logging, tracing, or alerting to services. Guides the three pillars of observability, structured logging standards, and instrumentation patterns. Not for language-specific logging setup (e.g., configuring a logger in Rust/Axum) — route those to the relevant language skill.

Use when building AI agents, designing LLM tool interfaces, working with pydantic-ai, pydantic-graph, or MCP. Guides agentic architecture, tool design, evaluation, and production patterns.

Use when writing Python code, configuring tooling, structuring projects, or working with FastAPI, httpx, pytest, or the Astral toolchain (uv, ruff, ty).

Use when writing Rust code, configuring Cargo workspaces, setting up clippy lints, designing error handling, or optimizing build profiles. Guides compiler-driven development and idiomatic Rust patterns. Excludes general Rust language questions without project tooling context (e.g., general async/await advice).

Use at the START of every new project inside a claude-worker VM. Describes the pre-baked SvelteKit 2 + PostgreSQL 16 environment that is already running — do not re-install Node, start a dev server, or set up a database.

Use after deploying any service to confirm it is live and healthy before marking a goal done. Covers HTTP health checks, Kubernetes pod status, and Playwright browser verification for UI apps.

Start a live dev server on port 8080 so doable shows an instant preview iframe. Use this whenever you have a working frontend that the user should be able to see before the full deployment cycle completes.

Use when creating, assigning tasks to, monitoring, or tearing down claude-worker autonomous agent VMs. Covers the full lifecycle via claude-ctl.

Use when running build, push, or deploy commands that need credentials (registry auth, API keys, tokens). Guides how to load credentials from the correct sources in this homelab.

Use when verifying user flows in a running web application using Playwright MCP tools. Covers navigation, form interaction, multi-step flows, and evidence collection.

Write documentation to Obsidian knowledge vault using MCP with proper frontmatter and structure

Use when writing human-readable documentation that should be stored in the Obsidian knowledge vault

Manage project documentation and planning in Obsidian vault using MCP