shermanhuman

Systematic debugging — reproduce, isolate, form hypotheses, instrument, fix, and add regression tests. Use when troubleshooting errors, failing tests, or unexpected behavior.

Analyzes task dependencies and groups independent steps into parallel batches. Use when executing multi-step plans or performing research across multiple sources.

Autonomous planning with internal reasoning. Researches, evaluates ideas, presents a plan. Minimizes user round-trips. Use before making non-trivial changes.

Reviews changes for correctness, edge cases, style, security, and maintainability with severity levels. 10 parallel checks with version-specific research. Use before finalizing changes.

Applies tests-first discipline (red/green/refactor) and adds regression tests for bugs. Use when implementing features, fixing bugs, or refactoring.

Mandatory checklist before claiming a task is done. Ensures verification, clean code, and accurate reporting. Use before saying "done" or "complete".

Reviews changes for correctness, edge cases, style, security, and maintainability with severity levels. 10 parallel checks with version-specific research. Use before finalizing changes.

Systematic debugging — reproduce, isolate, form hypotheses, instrument, fix, and add regression tests. Use when troubleshooting errors, failing tests, or unexpected behavior.

Semver release workflow — bump version, tag, push, verify CI. Use when releasing, bumping version, updating version, or when the user says 'release', 'bump', 'major', 'minor', 'patch', 'update version', 'new version', 'cut a release', 'tag', or 'goreleaser'.

Mise dev tool manager — installing tools, running mise-managed commands, and configuring .mise.toml. Use when installing tools, running CLI commands that aren't found, or setting up project environments.

Reference for the promptherder CLI — syncs AI agent rules, skills, and workflows from herd repos to agent targets. Use when installing, updating, or troubleshooting prompt distribution.

Docker best practices for multi-stage Alpine builds — security patches, image pinning, and layer optimization. Use when creating, reviewing, or updating Dockerfiles.

Tekmetric REST API — authentication, paginated endpoints, sync patterns, and undocumented behaviors. Use when integrating with Tekmetric shop management data (customers, vehicles, repair orders, employees, appointments).

DaisyUI v5 component library for Tailwind CSS — semantic UI classes, themes, layout patterns, drawer/sidebar gotchas, and component quick reference. Use when building, styling, or debugging UI with DaisyUI, Tailwind components, or DaisyUI themes.

Groq API syntax — Whisper transcription, audio processing, and prompt-based jargon guidance. Use when calling Groq endpoints or transcribing audio.

Core Phoenix framework patterns — context boundaries, LiveView, Ecto, and application architecture. Use when building, reviewing, or refactoring Phoenix applications.

Autonomous planning with internal reasoning. Researches, evaluates ideas, presents a plan. Minimizes user round-trips. Use before making non-trivial changes.

Reviews changes for correctness, edge cases, style, security, and maintainability with severity levels. 10 parallel checks with version-specific research. Use before finalizing changes.

Resolves target repository and storage location for conversation artifacts. Guarantees organized, time-sorted history.

Systematic debugging — reproduce, isolate, form hypotheses, instrument, fix, and add regression tests. Use when troubleshooting errors, failing tests, or unexpected behavior.

Analyzes task dependencies and groups independent steps into parallel batches. Use when executing multi-step plans or performing research across multiple sources.

Applies tests-first discipline (red/green/refactor) and adds regression tests for bugs. Use when implementing features, fixing bugs, or refactoring.

Mandatory checklist before claiming a task is done. Ensures verification, clean code, and accurate reporting. Use before saying "done" or "complete".

Ultra-compressed commit message generator. Cuts noise from commit messages while preserving intent and reasoning. Conventional Commits format. Subject ≤50 chars, body only when "why" isn't obvious. Use when user says "write a commit", "commit message", "generate commit", "/commit", or invokes /grugg-commit. Auto-triggers when staging changes.

Compress natural language memory files (CLAUDE.md, todos, preferences) into grugg format to save input tokens. Preserves all technical substance, code, URLs, and structure. Compressed version overwrites the original file. Human-readable backup saved as FILE.original.md. Trigger: /grugg:compress <filepath> or "compress memory file"

Quick-reference card for all grugg modes, skills, and commands. One-shot display, not a persistent mode. Trigger: /grugg-help, "grugg help", "what grugg commands", "how do I use grugg".

Ultra-compressed code review comments. Cuts noise from PR feedback while preserving the actionable signal. Each comment is one line: location, problem, fix. Use when user says "review this PR", "code review", "review the diff", "/review", or invokes /grugg-review. Auto-triggers when reviewing pull requests.

Ultra-compressed communication mode. Cuts token usage ~75% by speaking terse while keeping full technical accuracy. Supports intensity levels: lite, full (default), ultra, wenyan-lite, wenyan-full, wenyan-ultra. Use when user says "grugg mode", "talk like grugg", "use grugg", "less tokens", "be brief", or invokes /grugg. Also auto-triggers when token efficiency is requested.

Generator kinds, computed keys, template payloads, and Garage S3 credential patterns. Use when working on key creation, rotation, or template features.