{
  "markdown": "# skillsets\n\nA personal collection of agent skills for software and research engineering, packaged\nas a single cross-platform plugin. Each skill is a self-contained, model-agnostic\ninstruction set that runs on Claude (Claude Code), Codex, OpenCode, Antigravity\n(Gemini), and Hermes from the same source.\n\n## About\n\nEvery skill encodes a disciplined workflow rather than a prompt: classify the task,\ngather evidence, apply domain references, and verify before handoff. The goal is to\nkeep agent output grounded in explicit forces, trade-offs, and a verification path\ninstead of surface-level answers.\n\n## Skills\n\n| Skill | Scope |\n| --- | --- |\n| `skills/android-offsec` | Android offensive security, forensic extraction, mobile app exploitation, APK reversing, lock screen bypass, deeplink exploitation, rogue network attacks, deepfake social engineering, and mobile device investigation with chain of custody. |\n| `skills/software-architecture-engineering-uap` | Design, evaluate, document, modernize, and refactor software architecture and code structure; microservices patterns; clean code; IAM/auth; NFR tactics; ADR/RFC; implementation planning. |\n| `skills/uiux-frontend-engineering` | Research, design, implement, and validate UI/UX and frontend interfaces for web, mobile, responsive apps, design systems, theme tokens, and XR/spatial surfaces. |\n| `skills/data-architect-engineering` | Design, review, migrate, optimize, secure, and operate SQL, NoSQL, vector, graph, search, cache, big-data, lakehouse, and warehouse systems. |\n| `skills/device-driver-development` | Linux kernel device drivers, modules, device tree bindings, bus clients, DMA/IRQ/memory paths, bring-up, and upstreaming. |\n| `skills/realtime-systems-coding` | Concurrent, parallel, distributed, real-time, and network-dataplane systems; locks/monitors, RMS/EDF scheduling, DPDK/XDP, kernel bypass. |\n| `skills/tech-leadership` | CTO and technology leadership: strategy, cybersecurity, requirements, framework selection, people, delivery, risk, KPIs, roadmap, budgeting, and governance. |\n| `skills/academic-research-journal` | Evaluate, design, draft, and revise academic articles, literature/systematic reviews, journal submissions, and source/integrity audits. |\n| `skills/cyber-security-operations` | Cyber security operations: detection/hunting, DFIR/forensics, incident response, vulnerability management, supply chain security, DevSecOps, crypto/AI emerging threats, and offensive security. |\n\n## Layout\n\nThe repository is one plugin. Skills live under `skills/`, and each platform reads its\nown manifest:\n\n```\n.claude-plugin/\n  plugin.json         # Claude Code plugin metadata\n  marketplace.json    # Claude Code marketplace entry\n.codex-plugin/\n  plugin.json         # Codex plugin metadata + interface\n.opencode/\n  skills/             # OpenCode project-local skill mirror (symlinks to skills/)\nopencode.json         # OpenCode config and project instruction wiring\ngemini-extension.json # Antigravity (Gemini) extension manifest\nGEMINI.md             # Antigravity context file (includes AGENTS.md)\nAGENTS.md             # Cross-platform agent orientation\nhooks/\n  hooks.json          # SessionStart orientation hook (Claude Code)\n  session-start.sh    # emits the skillset orientation as additionalContext\n.mcp.json             # MCP servers (Context7, GitHub, Postgres, arXiv)\nLICENSE               # MIT\nskills/\n  <skill>/\n    SKILL.md            # Claude skill manifest (frontmatter: name, description)\n    agent.md            # Agent identity and operating model\n    agents/openai.yaml  # Codex interface config and default prompt\n    tasks/              # Step-by-step task workflows\n    references/         # Domain reference material\n    scripts/            # Helper scripts (static scans, probes)\n    commands/           # Slash commands (where applicable)\n```\n\n## Install\n\n### Claude (Claude Code)\n\n```\nclaude plugin marketplace add oghie/skillsets\nclaude plugin install skillsets\n```\n\nSkills under `skills/` auto-discover; Claude reads each `SKILL.md` and the\n`description` field drives auto-invocation.\n\n### Codex\n\nInstall the plugin from the same repository (`oghie/skillsets`). Codex reads\n`.codex-plugin/plugin.json`, which points at `./skills/` and exposes the bundled\nskills through its interface block.\n\n### OpenCode\n\nClone this repository and run OpenCode from the repo root:\n\n```\nopencode\n```\n\nOpenCode reads `AGENTS.md`, `opencode.json`, and the project-local\n`.opencode/skills/<name>/SKILL.md` paths. The `.opencode/skills/` entries are symlinks\nto the committed `skills/` directories, so Claude, Codex, OpenCode, and Antigravity use\nthe same source material.\n\n### Antigravity (Gemini)\n\nInstall this repository as a Gemini extension. Antigravity reads\n`gemini-extension.json` and loads `GEMINI.md`, which pulls in `AGENTS.md` for shared\nagent orientation.\n\n## MCP servers\n\nThe plugin ships a `.mcp.json` with four optional MCP servers. They are lazy-loaded\nand only matter for skills that benefit from live external data; each needs its own\nruntime and, where noted, configuration. A server that is not configured simply fails\nto start and shows up in `/mcp` — it does not block a session.\n\n| Server | Backs | Runtime | Configuration |\n| --- | --- | --- | --- |\n| `context7` | All engineering skills (current library/API docs) | Node 18+ (`npx`) | None required; set `CONTEXT7_API_KEY` for higher rate limits. |\n| `github` | tech-leadership, software-architecture (repos, issues, PRs) | none (remote HTTP) | OAuth on first use against `https://api.githubcopilot.com/mcp/`. |\n| `postgres` | data-architect-engineering (schema, EXPLAIN, index tuning) | `uv` (`uvx postgres-mcp`) | Optional. Set `DATABASE_URI` to your connection string to activate it; if unset it stays inactive and the plugin loads normally. Runs in `restricted` (read-only) mode. |\n| `arxiv` | academic-research-journal (paper search, source audits) | `uv` (`uvx arxiv-mcp-server`) | None required. |\n\nNotes:\n- `postgres` uses Postgres MCP Pro (`postgres-mcp`), not the archived\n  `@modelcontextprotocol/server-postgres`. `restricted` access mode wraps queries in\n  read-only transactions.\n- Install `uv` (https://docs.astral.sh/uv/) for the `postgres` and `arxiv` servers, and\n  Node 18+ for `context7`.\n- Remove any server you do not use by deleting its entry from `.mcp.json`.\n\n## Usage\n\nEach skill auto-surfaces by name and description. When a task matches a skill's\ndescription, the agent reads that skill's `skills/<name>/SKILL.md` or platform-local\nmirror and follows its workflow. All runtimes read the same `tasks/`, `references/`,\nand `scripts/`, so behavior stays consistent regardless of the host.\n\n### For newbies (install from scratch)\n\nIf you have none of these tools yet, pick a platform and follow two steps: install the\nCLI, then install skillsets.\n\n**Claude Code**\n\n1. Install the CLI:\n   - macOS / Linux: `curl -fsSL https://claude.ai/install.sh | bash`\n   - Windows (PowerShell): `irm https://claude.ai/install.ps1 | iex`\n   - Any OS with Node 18+: `npm install -g @anthropic-ai/claude-code`\n   - Check it works: `claude --version`\n2. Install skillsets:\n   ```\n   claude plugin marketplace add oghie/skillsets\n   claude plugin install skillsets\n   ```\n\n**Codex**\n\n1. Install the CLI:\n   - Any OS with Node 18+: `npm install -g @openai/codex`\n   - macOS (Homebrew): `brew install codex`\n   - Check it works: `codex --version`\n2. Install skillsets:\n   ```\n   codex plugin marketplace add oghie/skillsets\n   codex plugin install skillsets\n   ```\n   Or run `/plugins` inside Codex to browse and install it.\n\n**OpenCode**\n\n1. Install the CLI:\n   - macOS / Linux: `curl -fsSL https://opencode.ai/install | bash`\n   - Any OS with Node 18+: `npm install -g opencode-ai`\n   - macOS (Homebrew): `brew install anomalyco/tap/opencode`\n   - Check it works: `opencode --version`\n2. Use skillsets:\n   ```\n   git clone https://github.com/oghie/skillsets.git\n   cd skillsets\n   opencode\n   ```\n   OpenCode loads `AGENTS.md`, `opencode.json`, and `.opencode/skills/*`.\n\n**Antigravity**\n\n1. Download and install the Antigravity desktop app for Windows, macOS, or Linux from\n   https://antigravity.google, then sign in.\n2. Install skillsets as a Gemini extension:\n   ```\n   gemini extensions install https://github.com/oghie/skillsets\n   ```\n   Antigravity loads the repo's `gemini-extension.json` and `GEMINI.md`. Update later\n   with `gemini extensions update skillsets`.\n\n**Hermes**\n\n1. Install Hermes Agent:\n   - macOS / Linux: `curl -fsSL https://hermes-agent.nousresearch.com/install.sh | bash`\n   - Windows (PowerShell): `irm https://hermes-agent.nousresearch.com/install.ps1 | iex`\n   - Windows (desktop): download from https://hermes-assets.nousresearch.com/Hermes-Setup.exe\n   - Check it works: `hermes --version`\n2. Install skillsets:\n   ```\n   hermes skills tap add https://github.com/oghie/skillsets\n   ```\n   Then restart Hermes or run `/reload-skills`. Skills auto-surface by name and description.\n\n**After installing (all platforms)**\n\n- Restart the CLI or app so the skills load, then start a session and ask about one of\n  the skill domains; the matching skill loads on its own.\n- The optional MCP servers need extra runtimes — Node 18+ for `context7`, and `uv` for\n  `postgres` and `arxiv` (see [MCP servers](#mcp-servers)). Quick installs:\n  - macOS: `brew install node uv`\n  - Linux: Node from your package manager or nvm; uv via `curl -LsSf https://astral.sh/uv/install.sh | sh`\n  - Windows (PowerShell): install Node from nodejs.org; uv via `irm https://astral.sh/uv/install.ps1 | iex`\n\n## Conventions\n\n- Evidence over assertion: name the forces, trade-offs, and verification path before\n  recommending anything.\n- Version-sensitive facts (kernel APIs, library behavior) are verified against\n  current sources, not assumed.\n- `references/` is the source of truth per domain; `tasks/` orchestrates the workflow.\n- Shipped skills are self-contained and carry no external runtime dependencies.\n",
  "bytes": 9993,
  "sha": "1ee4c00dbcf180e2b7617e2773600b0433553fd9f280e7b83a52e95362a583b6",
  "repo_slug": "oghie/skillsets",
  "fonte": "repo",
  "truncated": false,
  "api": "https://agentalog.com/api/listings/plg_oghie_skillsets_4a5eccd3/readme"
}