dickwu

Full PIU API management toolkit — 57 MCP tools for managing projects, collections, requests, environments, data models, execution, search, sync, and OpenAPI generation. Use this skill whenever you need to interact with PIU — whether creating projects, sending API requests, managing environments, querying the knowledge graph, generating OpenAPI specs, or any task involving PIU data. Always activate when the user mentions PIU, API management, MCP tools, or any operation on API projects/collections/requests.

Deep sync a Hyperf PHP backend to PIU — extracts routes via Hyperf CLI, parses controller validation rules and response shapes, builds Eloquent model schemas, maps middleware to auth types, and creates PIU entities with full API documentation. Supports token-efficient incremental daily re-sync via manifest tracking. Use when the user says "sync hyperf", "import hyperf", "hyperf api sync", or when piu-backend-sync detects framework: "hyperf".

Analyze a backend repository and automatically import its API routes into PIU as projects, collections, requests, and environments via MCP tools. Use when the user says "sync backend", "import backend", "import API", "sync repo", "create project from repo", or provides a git URL with the word "analyze". Supports 14+ frameworks (Express, FastAPI, Django, Gin, Rails, Axum, Spring, NestJS, Hono, Echo, Fiber, Actix, Hyperf, Laravel) with automatic route extraction, data model creation, and version tracking. Also use when the user mentions importing routes, creating a PIU project from code, or converting a codebase to API documentation.

Analyze a frontend repository to discover all HTTP API calls and cross-reference them against a PIU project's backend routes. Use when the user says "sync frontend", "validate frontend", "check frontend API calls", "find API mismatches", "frontend API audit", "frontend coverage", or "API contract check". Identifies missing endpoints, undocumented calls, TypeScript contract violations, and can auto-fix by creating missing PIU requests or updating models. Supports re-sync with version tracking to detect changes since last audit. Also use when the user asks about API coverage, endpoint validation, or frontend-backend contract alignment.

Add (or rename, or change the shape of) an HTTP route in the nowledge axum service following the project's documented multi-file workflow. Use when the user says "add a route", "add an endpoint", "expose a new API", "wire up POST/GET/PUT/PATCH/DELETE /v1/...", or when changing an existing route's path/method/schema.

Run the project's standard pre-merge verification gauntlet for the nowledge crate (cargo fmt --check, cargo clippy --all-targets -- -D warnings, cargo check, cargo test) and report results. Use after substantive code changes, before claiming work is done, when the user says "verify", "check this", or "run the gauntlet", or when about to open a PR.

Run the gated Meilisearch integration test suite against the gfit host's Meilisearch through an SSH+socat tunnel. Use when asked to "run meili tests", "test against gfit meili", "run the meili integration suite", or when changes touch the Meili-backed repository path (src/meili.rs, src/repository.rs's MeiliKnowledgeRepository, or store hydration).

Set up a domain knowledge management system. Creates /knowledge/ directory with domain folders for tracking facts, hypotheses, and confirmed rules across sessions. Use when starting a new project, setting up knowledge management, or when the user mentions "mem-setup", "knowledge system", "domain rules", or "project memory".

Save domain knowledge insights after completing a task. Extracts facts, patterns, and observations from the current work and stores them in /knowledge/ domain folders. Promotes hypotheses to rules at 5+ confirmations, demotes rules when contradicted. Use when finishing a task, or when the user says done/finished/complete.

Auto-load domain knowledge at task start. Reads /knowledge/INDEX.md, finds relevant domain folders for the current task, loads rules.md (apply by default) and hypotheses.md (check if testable). Invoke manually when switching domains mid-conversation or when working in a domain you haven't loaded yet.

Deep inspection, interaction, debugging, and code review for Tauri v2 desktop apps. Use this skill whenever: working with a Tauri app's UI (clicking, filling forms, reading DOM, screenshots, dragging elements); debugging console logs, IPC calls, or Tauri events; reviewing component trees, accessibility, or visual regressions; testing user flows or validating IPC contracts; setting up tauri-connector in a new project. Also triggers on: DOM snapshots, element refs, webview interaction, drag-and-drop, IPC debugging, Tauri app testing, visual regression, admin/ front/ or tool/ desktop apps, @eN ref syntax, or any mention of tauri-connector CLI or MCP tools. This is Claude's bridge to any running Tauri v2 desktop app -- if a Tauri app is involved, use this skill.

Deep inspection, interaction, debugging, and code review for Tauri v2 desktop apps. Use this skill whenever: working with a Tauri app's UI (clicking, filling forms, reading DOM, screenshots, dragging elements); debugging console logs, IPC calls, or Tauri events; reviewing component trees, accessibility, or visual regressions; testing user flows or validating IPC contracts; setting up tauri-connector in a new project. Also triggers on: DOM snapshots, element refs, webview interaction, drag-and-drop, IPC debugging, Tauri app testing, visual regression, admin/ front/ or tool/ desktop apps, @eN ref syntax, or any mention of tauri-connector CLI or MCP tools. This is Claude's bridge to any running Tauri v2 desktop app -- if a Tauri app is involved, use this skill.

Publish a new release of the R2 Client desktop app and monitor CI. Use when the user asks to "publish a release", "cut a release", "bump the version", "release a new version", "check ci" after changes are committed, or "write release notes" / "publish release notes". Handles pre-flight checks, version bumping, git tagging, pushing, CI monitoring, and generating release notes from git diff using the project's publish.sh script and the gh CLI.

Cross-platform UI/UX design reviewer grounded in Apple Human Interface Guidelines principles. Use this skill to audit, review, critique, or improve any UI/UX design for mobile apps (iOS, Flutter, React Native) or desktop apps (macOS, Tauri, Electron). Triggers when the user mentions: design review, UI audit, HIG compliance, improving app design, design feedback, accessibility audit, or any request to check or improve a design against professional standards. Also use when the user uploads screenshots, mockups, wireframes, or design specs of a mobile or desktop app and wants feedback. Even if they just say "review my design" or "is this good UI", use this skill. Works for native and cross-platform frameworks including Flutter, Tauri, Electron, React Native, SwiftUI, and AppKit/UIKit.

Manage multiple workplaces (project directories) with multi-agent orchestration, isolated memory, and inter-agent communication. Use when the user mentions: workplace init/list/switch/scan/status/agents/export/import, managing projects, switching between codebases, multi-agent workflows, agent handoff, kernel agent, workspace structure, deploy environments, or any variation of "workplace" commands. Also triggered by /workplace slash command. Auto-detects .git folders as workplaces. Each workplace has its own agents, memory, skills, and deployment configs in a .workplace/ directory. Syncs context to Cursor, Claude Code, and OpenCode. Interactive Telegram/Discord UI with inline buttons for switching workplaces, starting agents, and deploying.

Drive the `gfit-cli` command-line client for the GFIT wellness API (`api.gfitwellness.ca`) confidently and safely. Use this skill whenever a task involves gfit-cli, GFIT coaching/admin data, or the GFIT API from the shell — listing or updating clients, coaches, check-ins, meal or workout plans, nutrition, foods, splits, metrics (weight / step / sleep / text logs), invoices, or Nowledge. It teaches the command model, how to discover any command and its parameters, authentication, the global flags (`--dry-run` / `--raw` / `--json`), arbitrary passthrough parameters, the response envelope, exit codes, and a safe workflow for write/destructive calls. Reach for it even when the user only names a GFIT operation without saying "gfit-cli" — e.g. "pull my clients", "who hasn't checked in this week", "list weight logs for user 42", "publish that meal plan", "search foods for chicken".

Publish a new release of the WordBrain desktop app and monitor CI on github.com/dickwu/wordbrain. Use whenever the user asks to "publish a release", "cut a release", "ship v0.x.y", "bump the version", "release WordBrain", "check ci" after a tag push, "write release notes", or "publish release notes". Also covers verifying the auto-updater endpoint, monitoring the Homebrew cask workflow, and recovering from common Tauri v2 release-pipeline failures (rustup `universal-apple-darwin`, frozen lockfile drift, missing `libpipewire`, signing-key format, Windows PowerShell expansion, `--no-default-features` rejected by tauri CLI, bash-3.2 globstar on macOS runners, Node 20 action deprecation, `release: published` not firing from `GITHUB_TOKEN`-created releases). Also covers the post-release symptoms: `brew install` failing with "undefined method 'license' for Cask" (cask DSL has no `license` stanza), the auto-updater failing with `None of the fallback platforms […] were found` (Tauri queries `darwin-aarch64`, not `dar

Drive the `source-map-php` Rust CLI to build, search, and manage Meilisearch-backed code indexes for Laravel and Hyperf PHP repositories. Invoke this whenever the user wants to index or reindex a PHP project, search indexed code (symbols, routes, tests, packages, schema), ask which tests cover a PHP symbol, run `source-map-php` subcommands (`init`, `doctor`, `index`, `search`, `validate`, `verify`, `promote`, `remove`), set up LLM-friendly code search for a PHP app, or troubleshoot Meilisearch-backed PHP code search — including casual phrasings like "index my laravel repo", "build code search for my hyperf service", "find routes for POST /foo in project X", "what tests cover App\\Services\\Foo::bar", "rebuild the php index after I added a class", or any mention of the `source-map-php` binary, Meilisearch + PHP code search, or the saved project registry at `~/.config/meilisearch/project.json`. Prefer this skill over ad-hoc `grep`/`rg` when the user wants structured, framework-aware search against an existing o

Auto-fix Tailwind CSS non-canonical class names across the project. Use when the user mentions tailwind canonical classes, suggestCanonicalClasses, fix tailwind classes, canonical fix, or wants to batch-fix Tailwind CSS class warnings like "h-[18px] can be written as h-4.5".

Publish a new release of iscsi-fuse. Bumps version, creates GitHub release, updates Homebrew formula, and installs locally via brew upgrade.