{
  "markdown": "# dot-agent 🤖\n\nA curated collection of agent skills for coding agents designed to enhance the workflows of Software Engineers (and, let's be honest, me primarily).\n\nSkills provide reusable, domain-specific knowledge and workflows that agents can use when relevant. Works with Cursor, Claude Code, Codex, OpenCode, Antigravity, GitHub Copilot, Amp, Roo Code, and other agent harnesses that support the Agent Skills standard.\n\n## What are Skills?\n\nSkills are modular, self-contained packages that extend agent capabilities with:\n- Specialized workflows for specific domains\n- Tool integrations and instructions\n- Domain expertise and best practices\n- Bundled resources (scripts, references, assets)\n\nSkills use **progressive disclosure** - keeping SKILL.md files under 200 lines with detailed content in `references/` directories loaded on-demand.\n\n## Getting Started 🚀\n\n### Install All Skills (Recommended)\n\nInstall all skills globally (to `~/.cursor/skills/`, `~/.claude/skills/`, etc. depending on agent):\n\n```bash\nnpx skills add siviter-xyz/dot-agent --global --agent cursor\n```\n\nBy default, `npx skills` will prompt before installing or updating skills. You can add `--yes` (or `-y`) to skip confirmations.\n\n### Install Specific Skills\n\nInstall individual skills by name:\n\n```bash\n# Install a single skill\nnpx skills add siviter-xyz/dot-agent --skill typescript --global\n\n# Install multiple specific skills\nnpx skills add siviter-xyz/dot-agent --skill typescript --skill python --skill context-engineering --global\n```\n\n### List Available Skills\n\nSee all available skills in this repository without installing:\n\n```bash\nnpx skills add siviter-xyz/dot-agent --list\n```\n\nTo discover skills across the broader ecosystem, use the built-in `skills find` command (powered by the internal `find-skills` helper skill; you don't need to install it separately):\n\n```bash\nnpx skills find dot-agent\n```\n\n### Install to Specific Agents\n\nTarget specific agent harnesses:\n\n```bash\n# Install to Cursor only\nnpx skills add siviter-xyz/dot-agent --agent cursor --global\n\n# Install to multiple agents\nnpx skills add siviter-xyz/dot-agent --agent cursor --agent claude-code --global\n```\n\n### Project-Level Installation\n\nInstall skills to your project (committed in repo):\n\n```bash\n# Install all skills to project\nnpx skills add siviter-xyz/dot-agent\n\n# Install specific skill to project\nnpx skills add siviter-xyz/dot-agent --skill psi\n```\n\nFor more options, see the [skills CLI documentation](https://github.com/vercel-labs/skills).\n\n### Sync from Local Repository\n\nWhen working on dot-agent locally, sync all skills to your local agent installation:\n\n```bash\n# Sync all skills from local dot-agent repo to Cursor\nnpx skills add \"$(pwd)\" --global --agent cursor\n\n# Sync to multiple agents\nnpx skills add \"$(pwd)\" --global --agent cursor --agent claude-code\n\n# Sync specific skills only\nnpx skills add \"$(pwd)\" --skill python --skill typescript --global --agent cursor\n```\n\nThis is useful when developing or modifying skills locally before publishing.\n\n## Available Skills\n\nSee [skills/README.md](skills/README.md) for a complete list of available skills.\n\n## Project-Specific Configuration\n\nFor project-specific rules and commands, commit them in your repository:\n\n- **AGENTS.md**: Project root or subdirectories for inline rules (works across agent harnesses, **PREFERRRED**)\n- **Rules**: `.cursor/rules/`, `.claude/rules/`, `.codex/rules/`, etc. in your repo\n- **Commands**: `.cursor/commands/`, `.claude/commands/`, `.codex/commands/`, etc. in your repo\n\nKeep project-specific content small and focused. Use skills for reusable knowledge.\n\n## Progressive Disclosure\n\nAll skills follow the progressive disclosure principle:\n\n- **SKILL.md** < 200 lines - Core principles and when to use\n- **references/** - Detailed documentation loaded on-demand\n- **scripts/** - Executable code for deterministic tasks\n- **assets/** - Files used in output (templates, etc.)\n\nThis keeps context windows clean while providing access to specialized capabilities when needed.\n\n## Contributing\n\nSkills are designed to be:\n- **Focused** - Each skill covers a specific domain\n- **Reusable** - Work across projects and teams\n- **Maintainable** - Clear structure with progressive disclosure\n\nWhen contributing:\n1. Keep SKILL.md under 200 lines\n2. Use references/ for detailed content\n3. Follow existing skill structure and the [Agent Skills specification](https://agentskills.io/specification)\n4. Prefer agent-harness agnostic skills; if a skill is environment-specific, document this clearly in its frontmatter and/or README\n5. For scripts inside skills:\n   - Prefer **uv scripts** (PEP 723 `# /// script` metadata with `#!/usr/bin/env -S uv run --script`) or **plain Bash** for simple glue\n   - Aim for cross-platform behavior (Linux, macOS; avoid hard-coding shell- or OS-specific paths)\n   - Recommend installing [`uv`](https://docs.astral.sh/uv/) locally and running scripts via `uv run` where appropriate\n6. Test installation and usage\n\n## References & Inspiration\n\n- [vercel-labs/add-skill](https://github.com/vercel-labs/add-skill)\n- [mrgoonie/claudekit-skills](https://github.com/mrgoonie/claudekit-skills)\n",
  "bytes": 5179,
  "sha": "562a9a0f4acdf311797114cf218ae1484a04980b8d1725082f3ba76e5eb0d1f7",
  "repo_slug": "siviter-xyz/dot-agent",
  "fonte": "repo",
  "truncated": false,
  "api": "https://agentalog.com/api/listings/skl_siviter_xyz_dot_agent_create_skill_048e19d3/readme"
}