{
  "skill": {
    "id": "nextcloud-search-and-read",
    "name": "nextcloud-search-and-read",
    "description": "Use when you need to search Nextcloud file names or content and then read the most relevant PDF, DOCX, MD, or TXT documents.",
    "path": "skills/research/nextcloud-search-and-read/SKILL.md",
    "tags": [
      "nextcloud",
      "search",
      "documents",
      "evidence"
    ]
  },
  "document": {
    "path": "skills/research/nextcloud-search-and-read/SKILL.md",
    "frontmatter": {
      "name": "nextcloud-search-and-read",
      "description": "Use when you need to search Nextcloud file names or content and then read the most relevant PDF, DOCX, MD, or TXT documents.",
      "version": "0.1.0",
      "author": "Hermes Agent",
      "license": "private",
      "metadata": {
        "hermes": {
          "tags": [
            "nextcloud",
            "search",
            "documents",
            "evidence"
          ],
          "related_skills": [
            "internal-deep-research"
          ]
        }
      }
    },
    "content": "---\nname: nextcloud-search-and-read\ndescription: Use when you need to search Nextcloud file names or content and then read the most relevant PDF, DOCX, MD, or TXT documents.\nversion: 0.1.0\nauthor: Hermes Agent\nlicense: private\nmetadata:\n  hermes:\n    tags: [nextcloud, search, documents, evidence]\n    related_skills: [internal-deep-research]\n---\n\n# Nextcloud Search and Read\n\n## Overview\n\nUse this skill to locate internal files and read the most relevant source documents before summarizing or synthesizing them.\n\n## Workflow\n\n1. Search by filename for likely documents.\n2. Search by content for key terms, names, or dates.\n3. Prioritize primary sources over summaries.\n4. Read the relevant PDF, DOCX, MD, or TXT files.\n5. Extract only the evidence needed for the task.\n\n## Best Practices\n\n- Search broadly, then narrow down\n- Keep track of source file names\n- Compare overlapping documents when claims matter\n- Prefer original source sections over secondhand notes\n\n## Verification\n\n- [ ] Relevant files identified by name or content\n- [ ] Primary sources opened\n- [ ] Evidence extracted from the actual document text\n- [ ] File names preserved in the working notes\n"
  }
}