ClaudeCode-global-config

Google Apps Script development best practices learned from production. Use when building Apps Script automation, payment API integrations, Google Sheets menu functions, modal dialogs, Poka Yoke quality gates, or running Apps Script functions from CLI. Covers UI dialog philosophy, API field discovery, submission architecture (source_amount vs transfer_amount), clasp CLI logging, run-appscript.sh function execution, and gate flow design.

Orchestration overview for the autonomous issue dispatch system. Handles bugs AND features from ANY entry point. References the component skills (dispatcher, handler) and per-project skills (bug-intake, qa-submission). Use when user says "dispatch", "fix these issues", "batch implement", "implement this", "build this", "develop this", "here's a plan", or any implementation/feature/fix request — including direct conversational requests to Claude Code.

Generalized Slack bug intake. Scans bug-report channels, auto-fixes small bugs, dispatches big work with clear specs, defers only genuinely ambiguous items to owner. Uses emoji reactions to track state. Each repo provides workspace config via local bug-intake skill. Supports shared channels where multiple repos receive bugs from one Slack channel via semantic routing.

PreToolUse hooks that block direct production database access and raw SQL write operations. Prevents bypass scripts that skip Drizzle ORM migration tracking, and HARD BLOCKS raw psql writes with no bypass — all writes must go through Admin API.

Standard development SOP for ALL code changes — bug fixes, features, refactors, any implementation. Defines the universal pipeline (investigate → fix → test → deploy → QA) and scope-based decision matrix. Also provides continuous scanning mode for autonomous operation. Use when implementing ANY code change, fixing bugs, building features, or when user says "start dev loop", "cowork", "fix this", "implement this", "build this". Each workspace can override with local skill.

Knowledge management for Claude Code projects. Use when: creating skills or agents, deciding where to put new knowledge (skill vs CLAUDE.md vs reference-data vs memory vs agent), routing information, CLAUDE.md hygiene, editing/modifying CLAUDE.md or skills or agents or memory files, extracting content from bloated files, or any discussion about knowledge architecture, documentation structure, or SOP placement. MUST be invoked BEFORE proposing any edits to CLAUDE.md, skills, agents, or memory - no exceptions.

Message drafting rules for emails, Slack, WhatsApp. Use when drafting messages on user's behalf, sending emails to company team, composing internal communications, or any message composition task. Covers draft review workflow, post-send disposition, company email addressing (role vs person), internal email format, and bilingual copypasta.

Google Drive CLI access via rclone. READ-ONLY for native Google formats (Sheets, Docs, Slides). Use for downloading, listing, searching Drive files.

Research methodology using Claude Code + Perplexity. Use when conducting competitive research, market analysis, person/lead research (pre-call prep, "who is this person", background checks), or any external research that requires verification. Core principle - Unknown is better than bullshit.

Context engineering audit after mistakes. Use when user says "retrospective", "retro", "introspect", "context audit", "what went wrong", or asks to analyze/prevent session mistakes. Invoke IMMEDIATELY when these triggers appear - don't do ad-hoc analysis.

ALL Slack operations. Use for ANY task involving Slack - messages, channels, members, workspaces, invites, permissions, history, reactions, threads, DMs, bots, apps, settings. If the word "Slack" appears in the request, use this skill.

End-of-session context preservation via GitHub Issues

Audit and suggest social media URL redirectors for your-domain.com. Scans the current redirector config (SSoT in next.config.mjs), cross-references with social/web presence discovered from the calling repo's context, and recommends new redirectors. Use when user says "check redirectors", "audit redirectors", "suggest redirectors", "social links audit", or wants to add new short URLs to your-domain.com.

Zero-tolerance SSoT enforcement across all repositories. Use when implementing ANY new code pattern, UI component, function, constant, or logic block. Triggers on all implementation work to enforce the mandatory pre-implementation grep check. Also use when reviewing code for duplication, discussing SSoT compliance, or when you notice yourself about to copy-paste code from one location to another.

Trello API operations — read cards, add comments, move cards, search boards, fetch board JSON. Use when any task involves Trello cards, Trello boards, Trello comments, student card lookups, or Trello data. Triggers — "Trello card", "update Trello", "check Trello", "add comment to card", "move card", "fetch board JSON", "Trello board export".

Systematic unbiased verification of completed work. Use when you need a fresh perspective to validate something you believe is correct. Spawns a new Claude instance with deliberately suspicious/neutral prompting.

YourBrand brand copywriting prompts and style guidelines. Use when generating tweets, LinkedIn posts, Facebook posts, or any text content for the YourBrand brand.

Granting access to repositories, shared resources, etc. Use when user asks to add collaborators, share repos, or modify access.

Extract and read previous Claude Code conversations from local storage. Use when reviewing past sessions, finding previous context, or debugging session history.

Self-contained delegation philosophy for handoffs that don't fall through cracks. Use when delegating tasks to team members, contractors, or AI agents.

Google Sheets browser automation pitfalls and workarounds. Use when automating Sheets via browser (clicking, typing, navigating cells). Not needed for rclone/CLI/MCP access.

Triage GitHub issues queue - check hygiene, identify actionable items, prioritize, invoke handler for each. The "what to work on" decision layer.

Diagnose a single GitHub issue - query logs, reproduce, collect evidence, hand off to CTO agent. The "how it broke" investigation layer.

Local Excel (.xlsx) file operations - safe editing with openpyxl, formatting preservation, spreadsheet creation from templates. Use when creating or editing local xlsx files, working with openpyxl, or creating new spreadsheets from Google Drive templates. Also covers rclone xlsx content search.

n8n workflow management via direct SQL (preferred over browser/chrome-agent). Use when modifying n8n workflows, debugging n8n issues, checking workflow executions, updating workflow nodes or connections, or when chrome-agent fails with n8n UI. Key insight - n8n stores workflows in PostgreSQL, direct SQL is faster and more reliable than UI automation. 🚨 CHROME AGENT IS FORBIDDEN FOR N8N - use SQL methods in this skill instead.

Base format for all report-style outputs (audits, triages, reviews). Defines item identification, severity sections, BLUF structure, and voice-friendly naming. Use when creating or modifying any skill that produces a report. Other report skills inherit from this.

Software installation preferences for Windows. Use when installing new tools, CLI utilities, or desktop applications. Covers winget vs UI installers, custom install locations, and package manager priority order.