{
  "markdown": "# AI1 Skills — SDLC Agent Skills for Python/React Projects\n\nA portfolio of **17 Agent Skills** covering the full software development lifecycle for **Python (FastAPI) + React/TypeScript** projects. Built on the [Agent Skills](https://agentskills.io) open standard — works with Claude Code, Cursor, GitHub Copilot, Codex, Windsurf, and other compatible tools.\n\n## Quick Start\n\n```bash\nnpx skills add hieutrtr/ai1-skills\n```\n\n| Starting a... | Guide |\n|---------------|-------|\n| **New project** from scratch | [Greenfield Guide](docs/greenfield.md) — step-by-step from planning to production |\n| **Existing project** adoption | [Brownfield Guide](docs/brownfield.md) — incremental adoption, security-first |\n\nTry these prompts after installing:\n\n```\n\"Plan the implementation for adding user authentication\"\n\"Create a FastAPI endpoint for user registration\"\n\"Review this code for security vulnerabilities\"\n```\n\nSee [Skill Composition Guide](docs/skill-composition.md) for how skills relate to each other.\n\n## Skills Overview\n\n| # | SDLC Phase | Skill | What It Does |\n|---|------------|-------|--------------|\n| 1 | Planning | `project-planner` | Feature breakdown, implementation plans, dependency mapping |\n| 2 | Planning | `task-decomposition` | Atomic task splitting, persistent task files, sizing criteria |\n| 3 | Architecture | `system-architecture` | Layer architecture, ADRs, database schema design |\n| 4 | Architecture | `api-design-patterns` | REST conventions, Pydantic v2 schemas, pagination, error format |\n| 5 | Implementation | `python-backend-expert` | FastAPI endpoints, repository pattern, SQLAlchemy 2.0, Alembic |\n| 6 | Implementation | `fastapi-patterns` | Middleware, dependency injection, WebSocket, JWT auth, lifespan |\n| 7 | Implementation | `react-frontend-expert` | Components, hooks, TanStack Query, forms, accessibility |\n| 8 | Testing | `react-testing-patterns` | Testing Library, MSW, hook testing, accessibility assertions |\n| 9 | Testing | `tdd-workflow` | Red-Green-Refactor enforcement for backend and frontend |\n| 10 | Testing | `pytest-patterns` | Fixtures, factories, async testing, mocking, parametrize |\n| 11 | Testing | `e2e-testing` | Playwright, page object model, auth reuse, CI integration |\n| 12 | Code Review | `code-review-security` | OWASP Top 10, SQL injection, XSS, secrets detection |\n| 13 | Code Review | `pre-merge-checklist` | Quality gates: linting, types, coverage, API compatibility |\n| 14 | Deployment | `deployment-pipeline` | CI/CD stages, canary rollout, rollback, GitHub Actions |\n| 15 | Deployment | `docker-best-practices` | Multi-stage builds, layer optimization, security, Compose |\n| 16 | Operations | `incident-response` | Severity classification, diagnostics, runbooks, post-mortems |\n| 17 | Operations | `monitoring-setup` | structlog, Prometheus, health checks, alerting, Sentry |\n\n## SDLC Flow\n\n```mermaid\nflowchart LR\n    subgraph Planning\n        PP[\"project-planner\"] --> TD[\"task-decomposition\"]\n    end\n\n    subgraph Architecture\n        SA[\"system-architecture\"]\n        AD[\"api-design-patterns\"]\n    end\n\n    subgraph Implementation\n        BE[\"python-backend-expert\"]\n        FP[\"fastapi-patterns\"]\n        FE[\"react-frontend-expert\"]\n    end\n\n    subgraph Testing\n        TDD[\"tdd-workflow\"]\n        PY[\"pytest-patterns\"]\n        RT[\"react-testing-patterns\"]\n        E2E[\"e2e-testing\"]\n    end\n\n    subgraph \"Code Review\"\n        SEC[\"code-review-security\"]\n        PMC[\"pre-merge-checklist\"]\n    end\n\n    subgraph Deployment\n        DOC[\"docker-best-practices\"]\n        DEP[\"deployment-pipeline\"]\n    end\n\n    subgraph Operations\n        MON[\"monitoring-setup\"]\n        IR[\"incident-response\"]\n    end\n\n    Planning --> Architecture --> Implementation --> Testing --> Code_Review[\"Code Review\"] --> Deployment --> Operations\n```\n\n## Installation\n\n### Option 1: skills.sh CLI (recommended)\n\nInstall the entire skill portfolio with one command:\n\n```bash\nnpx skills add hieutrtr/ai1-skills\n```\n\nThis downloads all 17 skills into your project's `.claude/skills/` directory.\n\nTo install a single skill:\n\n```bash\nnpx skills add hieutrtr/ai1-skills --skill \"python-backend-expert\"\n```\n\n### Option 2: Git clone\n\nClone the repository directly into your project:\n\n```bash\n# Project-scoped (this project only)\ngit clone https://github.com/hieutrtr/ai1-skills.git .claude/skills-repo\ncp -r .claude/skills-repo/skills/* .claude/skills/\nrm -rf .claude/skills-repo\n\n# Personal scope (all your projects)\ngit clone https://github.com/hieutrtr/ai1-skills.git ~/.claude/skills-repo\ncp -r ~/.claude/skills-repo/skills/* ~/.claude/skills/\nrm -rf ~/.claude/skills-repo\n```\n\n### Option 3: Manual copy\n\nDownload individual skill directories from [github.com/hieutrtr/ai1-skills](https://github.com/hieutrtr/ai1-skills) and place them in one of these locations:\n\n| Scope | Path | Applies To |\n|-------|------|------------|\n| Personal | `~/.claude/skills/<skill-name>/SKILL.md` | All your projects |\n| Project | `.claude/skills/<skill-name>/SKILL.md` | This project only |\n\n### Verify installation\n\nOpen Claude Code in your project and ask:\n\n```\nWhat skills are available?\n```\n\nClaude should list all 17 skills. You can also check context usage:\n\n```\n/context\n```\n\nExpected startup cost: 17 skills x ~100 tokens = ~1,700 tokens for Level 1 metadata.\n\n## Usage\n\n### How skills activate\n\nSkills use a **progressive disclosure** model with three levels:\n\n1. **Level 1 — Metadata** (~100 tokens per skill): Skill descriptions are loaded at startup so Claude knows what's available.\n2. **Level 2 — Full skill**: When Claude determines a skill is relevant to your request, it loads the complete `SKILL.md` into context.\n3. **Level 3 — References**: Supporting files (`references/`, `scripts/`) are loaded on-demand only when the active skill references them.\n\nYou don't need to do anything special. Claude activates the right skill based on your request.\n\n### Automatic activation\n\nJust ask Claude naturally. The skill descriptions contain phase-specific keywords that trigger the correct skill:\n\n```\n# Activates project-planner\n\"Plan the implementation for adding user authentication\"\n\n# Activates python-backend-expert\n\"Create a new endpoint for user registration\"\n\n# Activates pytest-patterns\n\"Write tests for the user service\"\n\n# Activates code-review-security\n\"Review this code for security vulnerabilities\"\n\n# Activates deployment-pipeline\n\"Set up the CI/CD pipeline for this project\"\n```\n\n### Direct invocation\n\nInvoke any skill directly with its name as a slash command:\n\n```\n/project-planner Add a payment processing module\n/python-backend-expert Create CRUD endpoints for orders\n/code-review-security Review the auth module\n/pre-merge-checklist Run all quality checks\n/tdd-workflow Implement the search feature using TDD\n```\n\n## How Skills Work Together\n\nSkills compose across SDLC phases — each phase produces artifacts consumed by the next.\n\n| Phase | Skills | Output |\n|-------|--------|--------|\n| Planning | `project-planner` → `task-decomposition` | Implementation plan → atomic task list |\n| Architecture | `system-architecture` + `api-design-patterns` | ADRs, layer decisions, API contracts |\n| Implementation | `python-backend-expert`, `fastapi-patterns`, `react-frontend-expert` | Backend + frontend code |\n| Testing | `tdd-workflow` + `pytest-patterns` / `react-testing-patterns` / `e2e-testing` | Test-driven features, E2E coverage |\n| Code Review | `code-review-security` → `pre-merge-checklist` | Security findings → quality gates |\n| Deployment | `docker-best-practices` → `deployment-pipeline` | Container images → CI/CD pipeline |\n| Operations | `monitoring-setup` → `incident-response` | Observability → runbooks |\n\nFor detailed composition rules, activation boundaries, and workflow diagrams, see:\n- **[Skill Composition Guide](docs/skill-composition.md)** — how skills activate, relate, and when *not* to use one\n- **[Greenfield Guide](docs/greenfield.md)** — full workflow for new projects (phases 1-8 in order)\n- **[Brownfield Guide](docs/brownfield.md)** — incremental adoption for existing projects\n\n## MCP Server Integration\n\nSkills become more powerful when paired with [MCP servers](https://modelcontextprotocol.io) that provide live data access:\n\n| Skill | MCP Server | What It Enables |\n|-------|------------|-----------------|\n| `project-planner`, `task-decomposition` | Jira MCP | Create issues, query backlog, link tasks |\n| `python-backend-expert` | PostgreSQL MCP | Schema inspection during implementation |\n| `pytest-patterns` | CI Server MCP | Trigger test runs, fetch coverage reports |\n| `code-review-security`, `pre-merge-checklist` | GitHub MCP | Fetch PR diffs, post comments, check CI |\n| `deployment-pipeline` | GitHub MCP, Kubernetes MCP | Trigger deploys, check pods, manage rollbacks |\n| `incident-response` | Datadog/PagerDuty MCP | Query metrics, check alerts, manage incidents |\n| `monitoring-setup` | Datadog MCP | Configure dashboards, verify alert rules |\n\n## Skill anatomy\n\nEach skill is a directory with this structure:\n\n```\nskill-name/\n├── SKILL.md              # Main instructions (required, <500 lines)\n├── references/           # Detailed docs, templates, examples (on-demand)\n│   ├── template.md\n│   └── patterns.md\n└── scripts/              # Executable validation/automation scripts\n    └── check.sh\n```\n\n### SKILL.md format\n\nEvery `SKILL.md` has YAML frontmatter and Markdown content:\n\n```yaml\n---\nname: python-backend-expert\ndescription: >-\n  Python backend patterns for FastAPI with SQLAlchemy 2.0, Pydantic v2,\n  and async patterns. Use during implementation when creating endpoints,\n  models, or services. Does NOT cover testing (use pytest-patterns).\nlicense: MIT\ncompatibility: 'Python 3.12+, FastAPI 0.115+, SQLAlchemy 2.0+, Pydantic v2'\nmetadata:\n  author: platform-team\n  version: '1.0.0'\n  sdlc-phase: implementation\nallowed-tools: Read Edit Write Bash(python:*) Bash(pip:*) Bash(alembic:*)\ncontext: fork\n---\n\n# Python Backend Expert\n\n## When to Use\n...\n\n## Instructions\n...\n\n## Examples\n...\n\n## Edge Cases\n...\n```\n\nKey frontmatter fields:\n- **`description`** — Controls when Claude activates the skill. Include phase keywords and negative keywords (\"Does NOT cover...\")\n- **`allowed-tools`** — Restricts what tools Claude can use. Planning skills get read-only access; implementation skills get write access.\n- **`context: fork`** — Runs the skill in an isolated subagent context\n\n## Target stack\n\nThese skills encode conventions for:\n\n| Layer | Technologies |\n|-------|-------------|\n| Backend | Python 3.12+, FastAPI 0.115+, SQLAlchemy 2.0+ (async), Pydantic v2, Alembic |\n| Frontend | React 18+, TypeScript 5+, TanStack Query 5+, Vite 5+, React Hook Form + Zod |\n| Testing | pytest + pytest-asyncio, Testing Library + Vitest, MSW 2+, Playwright |\n| Code Quality | ruff, mypy (strict), ESLint, Prettier |\n| Deployment | Docker (multi-stage), GitHub Actions, Prometheus, structlog, Sentry |\n\n## Context budget\n\nWith all 17 skills installed, the context impact is minimal:\n\n| Level | What loads | Token cost |\n|-------|-----------|------------|\n| Level 1 (always) | Skill descriptions for all 17 skills | ~1,700 tokens |\n| Level 2 (on activation) | 1-2 active SKILL.md files | ~5,000-10,000 tokens |\n| Level 3 (on demand) | Referenced files from `references/` | ~2,000-3,000 tokens |\n| **Typical total** | | **~10,000-15,000 tokens** |\n\nThis is under 8% of a 200K context window.\n\n## Customization\n\n### Override a skill\n\nTo customize a skill for your project, copy it to your project's `.claude/skills/` and edit. Project-scoped skills take precedence over personal-scoped ones.\n\n### Add project-specific conventions\n\nEdit the relevant `SKILL.md` to add your team's conventions. For example, add your database naming conventions to `python-backend-expert`, or your component library patterns to `react-frontend-expert`.\n\n### Extend with references\n\nAdd files to `references/` for large reference material. Reference them from `SKILL.md`:\n\n```markdown\nSee [API catalog](references/api-catalog.md) for endpoint documentation.\n```\n\nClaude loads these only when the skill is active and references them.\n\n## Compatibility\n\nThese skills follow the [agentskills.io](https://agentskills.io) core standard and work with:\n\n- [Claude Code](https://claude.com/claude-code)\n- [Cursor](https://cursor.sh)\n- [GitHub Copilot (VS Code)](https://code.visualstudio.com/docs/copilot/customization/agent-skills)\n- [OpenAI Codex](https://developers.openai.com/codex/skills/)\n- [Windsurf](https://windsurf.com)\n- Other agents supporting the Agent Skills standard\n\n## License\n\nMIT\n",
  "bytes": 12589,
  "sha": "92aff0f1a95ad058e7ff9fc5908cd9f19cd60974167854b5eb1da4641f8af7d5",
  "repo_slug": "hieutrtr/ai1-skills",
  "fonte": "repo",
  "truncated": false,
  "api": "https://agentalog.com/api/listings/skl_hieutrtr_ai1_skills_e2e_testing_1572e9a0/readme"
}