Agent skills for OpenAI Agents SDK

Cloud filesystem operations via Box CLI.

Personal boycott list \u2014 users can conversationally tell the agent\

TikTok content strategy, trend riding, hook writing, and viral mechanics.

Use when exploring a vague idea, finding inspiration, or researching a technique before planning.

Use before creative or constructive work (features, architecture, behavior).

Analyze §BRAND_NAME_TITLE§ agents logs

Extracts and structures application brand identities, USPs, and taglines by analyzing codebases and project documentation.

Scrape reviews, ratings, and brand mentions from multiple platforms using Apify Actors.

CTO-level architectural advisor for AI-native development.

Enforces a strict naming convention for getters to discourage encapsulation breakage and uphold the 'Tell Don't Ask' principle.

Analyze PR diffs and generate review comments from code quality, security, and performance perspectives.

Operates the browser via Playwright MCP — navigates, observes, interacts, and reports back based on any instruction.

Use when the user wants hyperflow's behavioral rules to apply outside the terminal CLI — in Claude Code Desktop, claude.ai web, or IDE…

Manages persistent work state (briefs) for local SEO engagements.

Diagnose and fix Bright Data common errors and exceptions.

Optimize Bright Data costs through tier selection, sampling, and usage

Install and configure Bright Data SDK/CLI authentication.

Configure Bright Data local development with hot reload and testing.

Execute Bright Data production deployment checklist and rollback procedures.

Implement Bright Data reference architecture with best-practice project

Apply production-ready Bright Data SDK patterns for TypeScript and Python.