{
  "id": 2021409,
  "title": "5 Portable Agent Skills for OpenCode and Claude Code",
  "url": "https://urgent.news/2026/08/19/5-portable-agent-skills-for-opencode-and-claude-code",
  "topic": "ai",
  "section": "AI",
  "published": "2026-08-19T21:19:48.000Z",
  "source": {
    "name": "Dev.to",
    "slug": "dev-to",
    "url": "https://dev.to/israelsaba/5-portable-agent-skills-for-opencode-and-claude-code-2a30"
  },
  "original_language": "en",
  "account": "1. The AI Writing Detector Skill evaluates English and Brazilian Portuguese text for signs that the content was generated by an AI. It checks for patterns such as repetitive sentence structure, filler words, excessive formatting, and em-dash usage. The Skill also provides a command-line interface (CLI) and a MCP server to lint text and files. It is useful for reviewing documentation, launch posts, landing pages, and changelogs before publication.\n2. The Text Counter Skill provides precise measurements of various elements in a text, including characters, words, sentences, paragraphs, lines, graphemes, bytes, and phrase occurrences. This is helpful when a rough estimate of the text length is not sufficient. Useful applications include writing a 155-character meta description for a package and ensuring its exact character count, reducing a GitHub issue title to 80 characters without removing error codes, and counting the occurrences of a specific phrase within a Markdown file.\n3. The HTML to PDF Skill transforms authorized public webpages into readable, browser-rendered PDFs. It utilizes Playwright and Chromium, waits for page assets, handles common cookie dialogs, removes ad blocks, and records the source URL and access time in each PDF footer. This Skill is useful for capturing public API documentation pages for offline review and creating readable PDFs of public release notes for research, design reviews, and internal reading packs.\n4. The ntfy Skill notifies users about the completion, failure, or blocking of agent tasks that may take longer than 30 seconds. It discourages the practice of embedding tokens in repositories, shell history, URLs, and notification content. The Skill supports both ntfy.sh and approved self-hosted ntfy servers. Useful prompts include notifying users once a lengthy export is finished and confirming that the notification token comes from the operating system's credential store.\n5. The Open Knowledge Format Skill generates source-linked Markdown research bundles containing an index, a research report, AI summaries, tables of evidence, and related output files. This Skill separates facts from interpretation and links claims back to their sources. Useful prompts include packaging research into an OKF bundle with source links and evidence rows, creating an ai_summary.md for another agent to continue the investigation, and structuring research in a way that it outlives a single chat session. This Skill, along with the others, enables a small, focused workflow for building with OpenCode or Claude Code.",
  "summary": "Agent Skills turn repeated prompts into reusable, inspectable workflows. This collection includes five small skills for work that comes up often when building with OpenCode or Claude Code: reviewing public copy, checking text limits, capturing public webpages as PDFs, sending task notifications, and structuring research for later use. The full index is available at Published Agent Skills . For…",
  "key_points": [],
  "editors_take": null,
  "illustration": null,
  "coverage": {
    "outlets": 1,
    "also_reported_by": []
  },
  "ai_generated": true,
  "disclaimer": "Summaries, key points and the editor’s take are written by software from other outlets’ reporting and may contain errors — always check the linked original."
}