eugenezhangco-spec
- 50 skills
- 0 followers
- 4 hours ago last updated
- ▌ Tdd Workflow · eugenezhangco-specUse this skill when writing new features, fixing bugs, or refactoring code. Enforces test-driven development with 80%+ coverage including unit, integration, and E2E tests.
- ▌ Backend Patterns · eugenezhangco-specBackend architecture patterns, API design, database optimization, and server-side best practices for Node.js, Express, and Next.js API routes.
- ▌ Frontend Patterns · eugenezhangco-specFrontend development patterns for React, Next.js, state management, performance optimization, and UI best practices.
- ▌ Verification Loop · eugenezhangco-specA comprehensive verification system for Claude Code sessions.
- ▌ Coding Standards · eugenezhangco-specUniversal coding standards, best practices, and patterns for TypeScript, JavaScript, React, and Node.js development.
- ▌ Blueprint · eugenezhangco-specTurn a one-line objective into a step-by-step construction plan for multi-session, multi-agent engineering projects. Each step has a self-contained context brief so a fresh agent can execute it cold. Includes adversarial review gate, dependency graph, parallel step detection, anti-pattern catalog, and plan mutation protocol. TRIGGER when: user requests a plan, blueprint, or roadmap for a complex multi-PR task, or describes work that needs multiple sessions. DO NOT TRIGGER when: task is completable in a single PR or fewer than 3 tool calls, or user says "just do it".
- ▌ Claude API · eugenezhangco-specAnthropic Claude API patterns for Python and TypeScript. Covers Messages API, streaming, tool use, vision, extended thinking, batches, prompt caching, and Claude Agent SDK. Use when building applications with the Claude API or Anthropic SDKs.
- ▌ Eval Harness · eugenezhangco-specFormal evaluation framework for Claude Code sessions implementing eval-driven development (EDD) principles
- ▌ Search First · eugenezhangco-specResearch-before-coding workflow. Search for existing tools, libraries, and patterns before writing custom code. Invokes the researcher agent.
- ▌ Deep Research · eugenezhangco-specMulti-source deep research using firecrawl and exa MCPs. Searches the web, synthesizes findings, and delivers cited reports with source attribution. Use when the user wants thorough research on any topic with evidence and citations.
- ▌ Security Scan · eugenezhangco-specScan your Claude Code configuration (.claude/ directory) for security vulnerabilities, misconfigurations, and injection risks using AgentShield. Checks CLAUDE.md, settings.json, MCP servers, hooks, and agent definitions.
- ▌ Docker Patterns · eugenezhangco-specDocker and Docker Compose patterns for local development, container security, networking, volume strategies, and multi-service orchestration.
- ▌ Agentic Engineering · eugenezhangco-specOperate as an agentic engineer using eval-first execution, decomposition, and cost-aware model routing.
- ▌ Continuous Learning · eugenezhangco-spec bundleAutomatically extract reusable patterns from Claude Code sessions and save them as learned skills for future use.
- ▌ Deployment Patterns · eugenezhangco-specDeployment workflows, CI/CD pipeline patterns, Docker containerization, health checks, rollback strategies, and production readiness checklists for web applications.
- ▌ AI First Engineering · eugenezhangco-specEngineering operating model for teams where AI agents generate a large share of implementation output.
- ▌ Enterprise Agent Ops · eugenezhangco-specOperate long-lived agent workloads with observability, security boundaries, and lifecycle management.
- ▌ Cost Aware LLM Pipeline · eugenezhangco-specCost optimization patterns for LLM API usage — model routing by task complexity, budget tracking, retry logic, and prompt caching.
- ▌ Agent Harness Construction · eugenezhangco-specDesign and optimize AI agent action spaces, tool definitions, and observation formatting for higher completion rates.
- ▌ Regex Vs LLM Structured Text · eugenezhangco-specDecision framework for choosing between regex and LLM when parsing structured text — start with regex, add LLM only for low-confidence edge cases.
- ▌
- ▌
- ▌
- ▌ Migrate · eugenezhangco-specMigrate to Athena from ANY previous AI setup — another Claude Code assistant, ChatGPT exports, Gemini history, Apple Notes, documents, or anything. Use when the user says 'migrate', 'bring over my old assistant', 'import my data', 'I have an existing assistant', 'I have ChatGPT history', 'I want to bring over my context', 'switch to Athena', or when the user mentions they have existing AI context they want to preserve. Also triggers when personal/me.md is empty and the user mentions having prior context. Handles everything: detects what they are coming from, imports what matters, sets up private infrastructure, gets the bot running. No re-onboarding needed if enough context is imported.
- ▌ Onboard · eugenezhangco-specOnboarding interview that turns a blank assistant into a personalized AI partner. Triggers on /onboard, first session, or when personal/me.md is empty. Also triggers on 'set up tools', 'connect my calendar', 'connect telegram', 'let's set up', 'connect my tools'. Do NOT trigger during normal conversations after onboarding is complete.
- ▌ Bookmark · eugenezhangco-specSave links, articles, videos, and resources to the user's Notion Bookmarks database with auto-detected source and an optional note. Use when the user shares a URL with context like "bookmark this", "save this link", "save for later", "add to bookmarks", "remember this link", or forwards content they want to revisit.
- ▌ Research · eugenezhangco-specSave research findings, articles, and learnings to the user's Notion Research database with key takeaways. Use when the user says "save this research", "add to research", "log this finding", "research note", or after a deep-research session when findings should be persisted.
- ▌
- ▌ Brain Dump · eugenezhangco-specCapture unstructured thoughts and ideas into the user's Notion Brain Dumps database. Splits multi-idea dumps into separate entries, auto-categorizes, saves as Raw. Use when the user says "brain dump", "capture this", "save this thought", "dump this", "note this down", or sends a stream of loosely connected ideas.
- ▌ Copywriter · eugenezhangco-specThe user's copywriter. Writes LinkedIn posts, Instagram captions, blog articles, emails, and any public-facing text in their voice. Adapts style by platform. Always loads voice-dna.md and runs the humanizer. Use when the user says 'write me a post', 'draft something', 'content session', 'write for LinkedIn', 'write for Instagram', 'blog post', 'help me write', or any request to create text as them.
- ▌ Transcribe · eugenezhangco-specDrop a video or audio URL — Athena fetches the audio, transcribes it, and returns a clean transcript plus a short summary. Works on YouTube, Instagram reels, TikTok, podcasts, and any yt-dlp-supported platform. Use when the user shares a link and wants the content extracted, not just bookmarked.
- ▌ Day Planner · eugenezhangco-specRuns structured planning sessions that map the user's day to their priority framework, delegation triage, goals-aware mentorship, and productivity principles. Two modes -- Evening (primary, plans tomorrow) and Morning (lightweight, confirms last night's plan). Use this skill whenever the user says 'let's plan tomorrow', 'help me structure my day', 'what should I focus on tomorrow', 'night-before planning', 'plan my day', 'help me prioritise tomorrow', 'what should I work on', 'build my schedule for tomorrow', 'how should I structure today', 'what's on my plate today', 'morning', or 'good morning'. Also activates when the user asks for a full daily plan across their priority areas. Do NOT use for creating individual calendar events (use calendar-manager). Output is a structured daily plan with time blocks, triage, gap analysis, and a mentorship push.
- ▌ Log Expense · eugenezhangco-specExtract and log an expense from a receipt photo into the user's Notion Expenses database. Auto-proposes the extraction, waits for approval before writing. Use when the user sends a receipt photo, says "log this expense", "add to expenses", "budget this", "track this spend", "save this receipt", or when an image sent on Telegram is clearly a receipt, invoice, or order confirmation.
- ▌ Video Ideas · eugenezhangco-specCapture video content ideas into the user's Notion Video Ideas database. Track from Draft to Ready to Published. Use when the user mentions a video idea, content concept, or says "video idea", "add to video ideas", "content idea", "I should film", "that would make a good video".
- ▌
- ▌ Self Upgrade · eugenezhangco-specSafe self-modification skill for upgrading Athena's own capabilities. Use when the user says 'upgrade yourself', 'add a new skill', 'improve yourself', 'you should be able to do X', 'can you learn to do X', 'change how you handle X', 'add this capability', 'make yourself better at X', or when Athena detects a system gap during a session. Also fires when the user wants to modify any file in .claude/rules/, .claude/skills/, or system files. Do NOT use for building user projects -- that is code-engineer. This skill is specifically for modifying Athena's own brain.
- ▌ Code Engineer · eugenezhangco-specWorld-class autonomous software engineer. Activates automatically whenever the user is building anything — fixing a bug, investigating an error, debugging, building a feature, creating an app, writing a script, making an automation, refactoring code, iterating on existing code, pasting an error message, pasting a stack trace, or saying 'build me X', 'make me X', 'I want an app that', 'something is broken', 'this does not work', 'help me with this code', 'fix this', 'why is this failing', 'automate this', 'create a tool for', or 'clean up this code'. Also activates whenever code in any language is being read, written, or modified. Runs the full engineering pipeline invisibly — the user describes what they want, Athena handles requirements, architecture, testing, security, and delivery. Do NOT replace building-explainer — both skills fire together during building mode.
- ▌ Weekly Review · eugenezhangco-specRuns a structured weekly review session where the assistant asks the user about their week, updates all context files, audits backlogs, and sets priorities for next week. Use this skill whenever the user says 'let's do the weekly review', 'weekly review', 'weekly check-in', 'let's review the week', 'end of week review', or 'how did this week go'. Also activates when the user wants to audit and clean up context files, update stale metrics, or align on priorities. Do NOT use for daily planning (use day-planner). This is the assistant and user sitting down once a week to get everything current.
- ▌ Calendar Manager · eugenezhangco-specManages all Google Calendar operations -- creating, reading, updating, and deleting events across the user's calendars. Use this skill whenever the user says 'add this to my calendar', 'create an event', 'block time for', 'what's on my calendar', 'check my schedule', 'when am I free', 'schedule a meeting', 'log a flight', 'add a reminder', or describes anything time-bound including meetings, calls, flights, errands, appointments, and travel plans. Also activates when the user mentions a specific date and time alongside a task or plan that needs to be tracked. Do NOT use for full structured day-planning sessions (use day-planner). Do NOT create events without the user explicitly asking -- present the proposed event first, create only after confirmation.
- ▌ Spending Summary · eugenezhangco-specQuery the user's Notion Expenses database and produce spending summaries — by month, category, merchant, tax-deductible status, or arbitrary date range. Use when the user says "how much did I spend", "summarize my spending", "budget for [period]", "tax-deductible total", "what did I spend on [category]", "monthly summary", "spending breakdown", or asks anything analytical about their expenses.
- ▌ Building Explainer · eugenezhangco-specExplains technical concepts in plain English using the Feynman Protocol -- name it, analogy, why it matters, stop. Use this skill automatically whenever a technical term appears that the user may not know, an error message needs translating, or the user says 'what does that mean', 'explain this', 'how does this work', 'why does this happen', 'what is a [technical term]', or 'I don't understand this'. Also fires automatically alongside code-engineer whenever building mode is active and any concept needs explaining. Do NOT use for writing, fixing, or reviewing code -- that is code-engineer. Do NOT replace code-engineer -- these two skills always run together, never instead of each other.
- ▌
- ▌ Update · eugenezhangco-specSave all session decisions, conclusions, and context to memory and STATUS.md so the team stays current across instances. Use when you say /update, "save this", "update yourselves", or at the end of any significant session.
- ▌ Humanizer · eugenezhangco-specRewrite existing text to remove AI patterns and match the project's voice. Use when pasting text from another AI, rewriting old drafts, or cleaning up imported content.
- ▌ E2e Testing · eugenezhangco-specPlaywright E2E testing patterns, Page Object Model, configuration, CI/CD integration, artifact management, and flaky test strategies.
- ▌ Security Review · eugenezhangco-specUse this skill when adding authentication, handling user input, working with secrets, creating API endpoints, or implementing payment/sensitive features. Provides comprehensive security checklist and patterns.
- ▌ Postgres Patterns · eugenezhangco-specPostgreSQL database patterns for query optimization, schema design, indexing, and security. Based on Supabase best practices.
- ▌ Database Migrations · eugenezhangco-specDatabase migration best practices for schema changes, data migrations, rollbacks, and zero-downtime deployments across PostgreSQL, MySQL, and common ORMs (Prisma, Drizzle, Django, TypeORM, golang-migrate).
- ▌ Iterative Retrieval · eugenezhangco-specPattern for progressively refining context retrieval to solve the subagent context problem
- ▌ Continuous Agent Loop · eugenezhangco-specPatterns for continuous autonomous agent loops with quality gates, evals, and recovery controls.