ChatGPT skills: how to create one, and 5 examples with live data
Blog post from Firecrawl
ChatGPT skills are reusable, folder-based workflows built around a SKILL.md file containing a name, trigger description, and instructions, allowing eligible Business, Enterprise, Healthcare, and Edu users to automate recurring tasks either by invoking a skill with @ or having ChatGPT select it based on a matching request. Skills can include scripts, templates, and reference files, use the open Agent Skills standard shared with Codex, and can be created through chat, an editor, uploads, or shared workspace installations, although Enterprise and Edu administrators control relevant permissions. Effective skills focus on one task, use specific trigger descriptions, define inputs and output formats, and identify any connected tools needed for live data. The material distinguishes skills from projects, which organize ongoing work, and custom GPTs, which are scheduled to retire in December 2026 and migrate into plugins, with custom actions requiring replacement apps. It also describes pairing skills with the Firecrawl plugin to search, scrape, crawl, and access Alexandria data providers such as SEC EDGAR, Sleeper, Apollo, Particle, Greenhouse, and Ashby, enabling workflows including fantasy lineup reviews, sales account briefs, public-company filing summaries, podcast sponsorship research, and hiring-signal reports, while noting that access, authentication, provider terms, and credit costs may apply.
Use this post, company, and trend context to find content marketing opportunities, perform competitive analysis, or address product feature gaps via the Plushcap MCP server or the Plushcap API.