dotclaude

Interactively scaffold or adapt a project-level extension for any plugin skill. Lists available skills, checks for existing extensions, and creates or opens the correct files. Triggers on: "extend skill", "customize skill", "skill extension".

Audit CLAUDE.md files for size, structure, and token efficiency. Checks both project-level and global CLAUDE.md. Warns when files exceed 25 lines (the recommended maximum for index-style CLAUDE.md). Suggests creating one if missing. Triggers on: "lint claude md", "claude md check", "check claude md", "CLAUDE.md zu lang", "audit claude md", "claude md audit". Do NOT auto-trigger from user phrasing for editing CLAUDE.md content or for /devops-setup-project. MAY be invoked explicitly via the Skill tool by other skills (devops-learn, devops-setup-project) as a post-edit gate.

Generate an interactive HTML page for analysis, plans, concepts, prototypes, comparisons, or creative work — open it in the browser and monitor user decisions (toggles, selections, comments) to feed them back into the workflow. Triggers on: "concept", "concept page", "interactive plan", "show me this as a page", "visualize this". Also auto-suggest when Claude completes analysis, planning, comparison, or concept work that would benefit from interactive decision-making. Do NOT trigger for: simple code explanations, debugging (use /devops-fix), or static documentation (use /devops-setup-readme).

Read session logs, runtime errors, and crash output to diagnose and fix the current issue — root-cause analysis first. Use when something is broken or behaving unexpectedly. Triggers on: "debug", "this is broken", "doesn't work", "error", "crash", "blank screen", "warum geht das nicht", "funktioniert nicht". Also triggers on pasted error/stack traces or when post.flow.debug hook detects repeated Bash failures. Do NOT trigger for general code questions.

Capture a long-term learning/correction and route it to the correct project- specific instructions (skill, skill-extension, deep-knowledge, or as a last resort CLAUDE.md) — NOT to personal feedback memory. Also prunes now-duplicate `feedback_*.md` entries with confirmation. Routing details live in the skill body (Step 5 table). Triggers ONLY on explicit invocation: "/devops-learn", "lerne das", "merk dir das fürs Projekt", "remember this for the project", "capture learning". Do NOT trigger for one-off conversational corrections or for personal feedback memory.

Manually update the devops plugin to latest from GitHub. Delegates to ss.plugin.update hook (pull + cache + registry), then adds changelog and verification report. Triggers on: "update plugin", "plugin updaten", "self update", "devops update", "neue version". Explicit user request only.

Analyze repository branch hygiene: unmerged branches, stale locals with deleted remotes, active sessions (worktrees), verify work landed in main. Results: interactive concept page with filters, 2-state delete controls, inline detail expand, and an Apply-Manifest + Dry-Run-Confirm before executing cleanup. Triggers on: "repo health", "branch cleanup", "branch hygiene". Explicit user request only.

Evaluate which agents are useful for a task and orchestrate their parallel or sequential execution. Use when the user explicitly wants orchestrated agent work instead of inline execution, or when a task clearly benefits from multi-agent collaboration. Triggers on: "agents", "run agents", "orchestrate", "use agents", "parallel agents", "multi-agent", "delegate to agents", "agent workflow". Do NOT trigger for: simple single-file edits, quick fixes, explanations, or research-only tasks (use the deep-research skill).

Fully autonomous agent orchestration for when the user is away from the PC. Runs agents without supervision (implementation, desktop interaction, live browser/app tests) and can OPTIONALLY SHUT DOWN THE PC after completion. Triggers: "autonomous", "run autonomous", "run this while I'm away", "afk mode", "autopilot". Do NOT trigger when the user stays present — use /devops-run-agents instead.

Milestone-centric backlog runner. Picks open GitHub milestones (or loose issues when there are none), then works the selected items off end-to-end — refine, implement, test/QA, and ship each one — unsupervised while the user is away or asleep, closing each issue and its milestone as it lands. Use when the user wants to work off the planned backlog / milestones without babysitting, or "let old ideas and fixes get worked off while I'm gone". Triggers: "backlog abarbeiten", "arbeite den backlog ab", "backlog runner", "run the backlog", "milestones abarbeiten", "arbeite die milestones ab", "arbeite den milestone ab". Do NOT trigger for maximizing token budget (that is /devops-run-burn, explicit-only), nor for a generic single ad-hoc AFK task (that is /devops-run-autonomous).

High-throughput autonomous task runner with aggressive parallelization. EXPLICIT INVOCATION ONLY — the user must type /devops-run-burn to activate. Do NOT trigger on any phrasing, keyword, or intent — not on "burn", "budget", "limit", "aufbrauchen", "maximize", "alles verbrauchen", "token", or ANY other wording. Only the literal /devops-run-burn command.

Create GitHub issues with enforced title format, labels, and optional milestone and project board integration. Also handles milestone creation and naming. Use when the user wants to create a GitHub issue, plan a milestone, or manage issue lifecycle. Triggers on: "neues Issue", "create issue", "Issue erstellen", "mach ein Issue", "new issue", "milestone planen", "plan a milestone". Do NOT trigger for: PR creation (use /devops-ship), commit operations (use /devops-commit), or code implementation.

Audit or initialize a project's repository hygiene: .gitignore, LICENSE, README, .editorconfig, .gitattributes, and AI tooling config. Also scaffolds plugin skill extensions for the project. Triggers on: "set up this project", "init repo", "audit gitignore", "add license", "fix gitignore", "repo hygiene". Do NOT trigger for README generation (/devops-setup-readme), CLAUDE.md edits, or source code changes.

Generate a polished, modern README.md for any project. Use when the user explicitly calls /devops-setup-readme, AND when Claude is about to create, rewrite, or substantially update a README.md. Triggers on: "create a readme", "update the readme", "improve the readme", "README erstellen", "README aktualisieren". Do NOT trigger for minor one-line edits like bumping a version number.

Full end-to-end shipping pipeline using MCP tools: ship_preflight, ship_build, ship_version_bump, ship_release, ship_cleanup, render_completion_card, then silent memory consolidation. Supports hierarchical merges (sub-branch → feature → main). Use when work is ready to land. Triggers on: "ship it", "push and merge". Do NOT trigger during coding/debugging or for commits without shipping.

Stabilization pass: run the full test suite, fix bugs autonomously, identify architecture smells, write regression + missing-coverage tests, apply consistency fixes (spacing, tokens, typography, icons, colors, state-visuals) across the codebase. Does NOT introduce new UI elements, new buttons, re-arrangements, or substantial position changes — those belong to /devops-tune-polish. Triggers on: "harden", "stabilize", "härten", "stabilisieren", "bug pass", "consistency pass", "lint und fix". Skips all confirmations when invoked with --autonomous. Do NOT trigger for: feature work, new UI structure, theme changes.

UI refinement pass: visual consistency (spacing, tokens, typography, icons, colors), state-visuals (hover/focus/disabled/loading/empty/error), UI-side functionality verification (clicks react, forms validate, loading shows), and small backend fixes when they are demonstrably UI-related (lag, missing field, broken contract). May introduce structural UI changes (new buttons, repositions, re-arrangements) but ALWAYS with user approval. Triggers on: "polish", "ui polish", "ui angleichen", "design konsistenz", "feinschliff", "visuell aufräumen", "design pass". Skips approval prompts when invoked with --autonomous (still respects the "structural changes flagged not applied" rule). Do NOT trigger for: backend-only work, feature implementation, theme/style overhaul.

Strategic reset for stuck development: iterations circle without reaching the goal, or a named part of the app (or the whole app) is functionally and visually not where it should be and incremental fixes stopped helping. Re-derives the original goals from repo + live-app evidence, validates them in a short question round (incl. jointly calibrating how much may be torn down), generates genuinely fresh approaches via code-blind lens agents, reconciles them against the codebase, lets the user decide on a devops-concept page, then hands the chosen approach to devops-run-autonomous for implementation with test-plan verification (deep-knowledge/test-plan.md). Triggers on: "festgefahren", "stuck", "unstuck", "wir drehen uns im Kreis", "neu denken", "rethink", "frischer Ansatz", "fresh approach", "komplett neu denken", "das führt zu nichts". Do NOT trigger for: debugging/errors (use /devops-fix), incremental consistency or UI passes (/devops-tune-harden, /devops-tune-polish), or normal feature work.

Reclaims orphaned Claude Desktop MCP server processes leaked by previously-closed sessions, in the background, without a restart.

Project ship extension for the dotclaude plugin-source repo. After a successful ship to main, syncs the locally-installed devops plugin to the just-shipped version so this Claude Code instance doesn't end up on a stale install.

Codebase knowledge graph via the external graphify CLI, default-on and opt-out. Detects graphify, auto-installs it in the background if missing, keeps the graph fresh with a key-less `graphify update .`, and answers codebase questions with `graphify query`. Enabled automatically in every project unless `.claude/graphify.json` (or the global `~/.claude/graphify.json`) has `{"consent":false}`. Once enabled, the graph is kept fresh windowlessly (SessionStart refresh + PreToolUse self-heal — graphify's own git hooks are removed, not installed) and broad raw-file searches are hard-gated toward the graph. Deliberately does NOT install graphify's own PreToolUse hook (would collide with the devops token-guard); the enforcement is devops-owned. Triggers: "knowledge graph", "graphify", "code graph", "/devops-graph". Do NOT trigger for simple single-file lookups.

Deliberate channel promotion for the ring model: move an already-shipped version alpha→beta→stable (or fast-track alpha→stable) by re-tagging the SAME commit via the ship_promote MCP tool. Never rebuilds, never bumps versions, never runs autonomously — promotion is always a user decision. Triggers on: "release", "promote", "promotion", "channel release", "auf stable heben", "promote to beta", "promote to stable". Do NOT trigger for shipping new work (use /devops-ship) or plugin updates (/devops-plugin-update).

Fetch live token usage for completion card battery line. Reads the native statusLine-written usage file first (no scrape), falls back to the Edge CDP scraper or cached data. Run silently pre-card, or manually: "refresh usage", "wie viel hab ich verbraucht", "token budget".

Interactive setup for the local-llm plugin — ask the user for their AnythingLLM API key, save it user-globally, and verify the connection. Use when the SessionStart hook reports `phase: needs_api_key`, `auth_failed`, or when the user explicitly asks to configure or reconfigure the local LLM. Triggers on "local llm setup", "anythingllm key", "local-llm configure".

Create a conventional commit for staged/changed files — enforces type, scope, co-author footer. Handles amend for "update/fix last commit". Do NOT trigger for push, PR, or ship operations.

Stop-hook self-audit and learning loop — review feedback, internalize skills, calibrate behavior.

Session-start hook that checks for uncommitted/unpushed changes and warns the user only when issues exist.