This document defines the patterns and workflows for working with the shared-gha repository.

>-

**Project**: MacroEconomic Uncertainty database replication (Comunale & Nguyen 2025)

You are Session Scribe, a specialized Claude skill for observing and documenting Claude Code development sessions.

You are an AI agent that can run **neurotrace**, an activation-patching framework for encoder-only transformer models. Use this skill when a user wants to understand where factual knowledge is stored

name: isf-grant

DFS graph crawler + full frontend audit. Playwright crawls the live app as a graph (pages=nodes, links=edges), DFS from entry, audits EVERY reachable node: multi-breakpoint screenshots (320/768/1440/1920px), WCAG 2.1 AA (axe-core), state verification (loading/error/empty/success), dark mode, interactive states, performance (LCP/CLS/FCP), visual regression, console errors, API validation, hydration. Auth support (storageState), CI mode (exit 1 on CRITICAL, text-only), max-pages cap, crawl failure recovery. Auto-detects 7 frameworks: Next.js, Vite/React, Vue/Nuxt, SvelteKit, Astro, Remix, Angular. Actions: test, crawl, audit, screenshot, verify, scan, regress, diff. Outputs: HTML graph report, topology map, per-node pass/fail, severity-rated issues.

AI video production workflow using Remotion. Use when creating videos, short films, commercials, or motion graphics. Triggers on requests to make promotional videos, product demos, social media videos, animated explainers, or any programmatic video content. Produces polished motion graphics, not slideshows.

**Scope of this file:** Design decision quality, UX principles, and the "why" behind UI choices.

SpaceMolt Agent Skill

>

This skill should be used when debugging web applications, diagnosing page errors, inspecting console output, or capturing screenshots of pages. It provides Chrome DevTools Protocol (CDP) automation via the chromectl.py script for collaborative or automated browser debugging.

This document lists every best-practice rule to follow for this project going forward.

This skill activates for web scraping and Actor development. It proactively discovers APIs via traffic interception, recommends optimal strategy (traffic interception/sitemap/API/DOM scraping/hybrid), and implements iteratively. For production, it guides TypeScript Actor creation via Apify CLI.

Audits and optimizes websites for search engine visibility (SEO) and AI search citation (GEO), covering technical health, E-E-A-T content scoring, domain authority, structured data, rich results, and entity signals. Use when running SEO audits, diagnosing traffic drops or ranking losses, generating Schema.org JSON-LD, checking Core Web Vitals, crawlability, robots.txt, sitemaps, hreflang, backlinks, planning content strategy or site migrations, fixing indexing issues, or optimizing for AI Overviews, ChatGPT, and Perplexity. NOT for paid ads (PPC/SEM), social media strategy, email marketing, or general web development unrelated to search.

Create, edit, export, and review draw.io diagrams. Use for native .drawio XML generation, PNG/SVG/PDF export, SVG overlap, border-overlap, label-intrusion, label-rect, short-terminal, and text-overflow linting, layout adjustment, and AWS icon usage.

>-

Use when a task involves many files (>10), the total content exceeds comfortable context size, or the task benefits from divide-and-conquer. Runs an agent that discovers files via shell tools, peeks strategically, and recursively decomposes into sub-agents. Bundles bashrlm context guards — output from high-output commands (cat, grep, rg, jq, find, ls, curl, etc.) is automatically truncated so the agent never floods its own context window.

>

Launch Claude Code async in background with automatic delivery to Telegram/WhatsApp. Use for coding, refactoring, codebase research, deep research, comprehensive investigations, multi-source reports, file generation, and complex multi-step automations. NOT for quick one-off questions or real-time interactive tasks. Includes strict thread-safe routing + E2E operator validation workflow.

You are building a plugin for **Teleton**, a Telegram AI agent on TON. Ask the user what plugin or tools they want to build, then follow this workflow.

Create precise, programmatic animations for technical blogs, educational content, and concept visualization using Manim Community Edition (ManimCE).

memwal

Karpathy's LLM Wiki pattern, built for Claude Code. Long-term memory for Claude via Obsidian: captures URLs, PDFs, GitHub, YouTube and compiles them into a structured Obsidian vault with citations that grows smarter every session. Second brain, zettelkasten, and autonomous research assistant in one. Keywords: llm-wiki, obsidian-plugin, knowledge-base, compounding-knowledge, claude-code, pkm, llm-memory.

Universal techniques for crafting effective prompts across any LLM.

libdestruct is a Python library for destructuring binary data into typed objects. It maps raw bytes to C-like types (integers, floats, strings, structs, pointers, arrays, enums, bitfields) with read/w

When bumping a patch version (e.g., `1.9.0` -> `1.9.1`), follow this checklist.

Create a SOLIDWORKS (CAD) desktop add-in using the SOLIDWORKS API SDK templates or by implementing ISwAddin/SwAddin manually; includes COM/registry registration, ConnectToSW/DisconnectFromSW lifecycle, CommandManager UI, callbacks, icons, and toolbar considerations.

Wrapping manager delegations in local breadcrumbs for cross-context operation tracking when both servers are installed

Analyze political, economic, social, technological, environmental, and legal forces. Use when external market shifts could materially affect a product, roadmap, or strategy.

Configure Slack channel tokens (bot token + app-level token)

>-

Use when asked to review a PR, or when /review is invoked with a PR number or URL. Performs a focused code review checking for bugs, security, performance, and test gaps, then posts findings as a PR comment and formal GitHub review.

Language-agnostic coding principles for maintainability, readability, and quality. Use when implementing features, refactoring code, or reviewing code quality.

Configures Laravel Nightwatch data collection, sampling rates, filtering rules, and redaction policies. Use when setting up Nightwatch, managing data volume, protecting sensitive data (PII), or optimizing event collection for production workloads.

9-language quality gate validation: linting, formatting, type checking, and test execution. Validates code changes meet quality standards before completion. Use when: (1) after code implementation, (2) before PR creation, (3) as part of /orchestrator Step 6, (4) manual quality check. Triggers: /gates, 'quality gates', 'run validation', 'check quality', 'validate code'.

Direnv environment management for automatic per-project shell configuration. Use when setting up .envrc files, configuring project-specific environment variables, or integrating direnv with development tools like nix, asdf, pyenv, or nvm.

Systematic web application debugging using Chrome DevTools MCP and Playwright MCP with intelligent validation and app-specific context discovery. Use for debugging web apps, APIs, authentication flows, and UI issues.

Maven build configuration for the crypto-scout multi-module Java 25 project

Apply when creating or editing INFO, SPEC, IMPL, TEST, FIX documents, or STRUT plans

Execute work items based on their assigned mode (autopilot, confirm, validate). Supports single-item and multi-item (batch/wide) runs.

>-

Guide REPL-driven development workflow for Clojure projects. The foundational approach is "tiny steps with high quality rich feedback" — rapid iteration with immediate validation before committing cod

Personal assistant operating routines (brief, inbox digest, auth health, daily planning, lightweight heartbeat alerts) with low token overhead.

Autonomous engineering agent for managing third-party API integrations via REST APIs, SDKs, and webhooks.

Use when developing applications or components using TypeScript, ensuring type safety and leveraging TypeScript features.

Build Keynote-style single-file HTML slide decks with brand-ready templates, minimal navigation, and Gemini nano banana media generation. Includes Narrative Engine integration for framework-driven deck creation with 17 proven storytelling structures and 5-agent review panel. Use when creating or editing slide decks, transforming content into presentations, or generating slide visuals.

Update existing content with fresh information and improvements.

Creates feature dashboards in Hex Threads. Given a feature name (and optionally a PRD, Figma link, or context), discovers events, builds a metric spec, gets human approval, and instructs Hex to build charts using business context. Hex writes its own SQL — Claude provides context, table/event names, and exact segmentation logic. Use when someone asks to create, build, or generate a feature dashboard.

Scan lyrics for pronunciation risks, prevent Suno mispronunciations