Use when you want to evolve your own capabilities — create new tools to avoid repeating patterns, add hooks to improve your own behavior, build skills to accumulate domain knowledge, or override built-in tools to adapt to project needs. This is your self-iteration interface.
Turn a design reference into matching, production code and keep code in sync with the design. Ingest a screenshot, Figma export, design-tokens file, or a design URL; extract colors/spacing/type/radii; map them onto the project's existing design-system tokens (reuse, don't duplicate); generate or update components; then verify visual parity with a screenshot. RTL/Hebrew-aware. Activate for: design to code, sync design, match this design, implement this Figma, pixel-perfect, design tokens, design handoff, עיצוב לקוד, סנכרון עיצוב, התאמה לעיצוב, פיגמה.
Semantic codebase search & indexing (Cursor-style). When a question is conceptual ('where is X handled', 'find everything related to auth') or the codebase is large/unfamiliar, index it once with the code-memory tools and use semantic search instead of blind grep — then verify hits by reading the real files. Activate for: where is this implemented, find where, search the codebase, semantic search, index the codebase, find all usages of, how does X work, איפה זה מטופל, חפש בקוד, אינדקס לקוד, חיפוש סמנטי, מצא את כל המקומות.
Deploy / sync the current LOCAL workspace to its linked REMOTE host over SSH, and manage that remote (create dirs, run commands, restart services) — all from the local workspace. Use when the user asks to deploy, push to production, sync to the server/remote, ship, release, or "handle the remote". Reads the per-workspace link from .squadcoder/remote.json. Security-first: never ships secrets, dry-runs first, confirms destructive actions. Auto-fires on: deploy, push to production, ship it, release, sync to remote, deploy to relay, manage the server, פרוס, העלה לשרת, דחוף לפרודקשן.
Review a diff or PR like a senior engineer: correctness and edge cases first, then security, performance, readability, and tests — with severity-tagged, actionable findings tied to specific lines. Prioritizes real bugs over style nits and verifies claims against the code. Activate for: review this code, code review, review my PR, review the diff, find bugs, security review, סקירת קוד, בדיקת קוד, ביקורת קוד.
Design and build a production MCP (Model Context Protocol) server: pick stdio vs HTTP transport, define tools/resources/prompts with strict input schemas, handle auth and secrets safely, and ship it. Covers Node/TypeScript and Python SDKs, error handling, and how to register it in an opencode/Claude/Codex config. Activate for: build an MCP, MCP server, model context protocol, add a tool server, expose an API as MCP, MCP טוול, שרת MCP.
Test a web app end-to-end with Playwright: drive real flows, assert on visible state, catch console/network errors, and do visual + RTL/Hebrew checks. Covers writing resilient selectors (roles/text over CSS), waiting on conditions (never sleeps), screenshot diffing, and a quick smoke pass before shipping. Activate for: test the web app, e2e test, playwright, browser test, check the UI, visual regression, smoke test, בדיקת אתר, בדיקות אוטומציה, בדיקת ממשק.
Design banners for social media, ads, website heroes, creative assets, and print. Multiple art direction options with AI-generated visuals. Actions: design, create, generate banner. Platforms: Facebook, Twitter/X, LinkedIn, YouTube, Instagram, Google Display, website hero, print. Styles: minimalist, gradient, bold typography, photo-based, illustrated, geometric, retro, glassmorphism, 3D, neon, duotone, editorial, collage. Uses ui-ux-pro-max, frontend-design, ai-artist, ai-multimodal skills. Activate for: באנר, עיצוב באנר, באנר לפייסבוק, באנר לאינסטגרם, קריאייטיב, מודעה, פרסומת, תמונת נושא, עיצוב גרפי.