All authors
Claude Skills by Fqih
github.com/Fqih34 skills0 installs74 views
- Agent DesignUse when designing AI agents — tool use patterns, ReAct/plan-and-execute loops, memory, multi-agent orchestration, error recovery, observability. Reference LangGraph, CrewAI, Anthropic tool use. Trigger on "agent", "tool use", "react", "autonomous", "multi-agent", "/agent-design".Votes: 0GitHub stars: 2
- Api DocsUse when generating API documentation — OpenAPI/Swagger auto-generation from code, MkDocs Material, Docusaurus, Mintlify, Postman/Insomnia export. Trigger on "openapi", "swagger", "api docs", "mkdocs", "docusaurus", "/api-docs".Votes: 0GitHub stars: 2
- Atomic Github PushUse when committing and pushing to GitHub — guarantees every commit passes lint/format checks, has no Claude co-author or generated-with footer, uses the user's real identity (Fqih / mhmdfkih21@gmail.com), and lands on the remote as a single all-or-nothing unit. Trigger on "commit and push", "push to github", "atomic push", "/atomic-push".Votes: 0GitHub stars: 2
- Audit WorkflowUse when ending a sprint/cycle — run comprehensive code audit (security, quality, tests, dead code, deps, docs) and post findings as GitHub issues for next-cycle pickup. Trigger on "audit", "sprint review", "end of cycle", "/audit-end-sprint".Votes: 0GitHub stars: 2
- Backend PatternsBackend architecture patterns, API design, database optimization, and server-side best practices for Node.js, Express, and Next.js API routes.Votes: 0GitHub stars: 2
- Clickhouse IoClickHouse database patterns, query optimization, analytics, and data engineering best practices for high-performance analytical workloads.Votes: 0GitHub stars: 2
- Coding StandardsUniversal coding standards, best practices, and patterns for TypeScript, JavaScript, React, and Node.js development.Votes: 0GitHub stars: 2
- Continuous LearningAutomatically extract reusable patterns from Claude Code sessions and save them as learned skills for future use.Votes: 0GitHub stars: 2
- Dependency AuditUse when auditing third-party dependencies — known CVEs, outdated packages, supply chain risk, SBOM generation, license compliance. Trigger on "pip-audit", "npm audit", "vulnerability", "CVE", "SBOM", "/audit-deps".Votes: 0GitHub stars: 2
- Docker PatternsUse when writing Dockerfiles, docker-compose, containerization — multi-stage builds, layer caching, security, image size optimization, health checks, networks, volumes. For Avo sandbox + general. Trigger on "dockerfile", "docker-compose", "container", "image", "build context", "/docker-check".Votes: 0GitHub stars: 2
- Document IntelligenceUse when reading, extracting, summarizing, or citing from PDF / Word / Markdown / HTML / EPUB documents. Triggers on "read this pdf", "extract from word", "summarize paper", "cite from document", "/pdf-read", "/doc-extract", "/doc-summarize".Votes: 0GitHub stars: 2
- Ds WorkflowUse when running data science experiments with mlflow — initialize tracking server, log params/metrics/artifacts, compare runs, register models, reproduce experiments. Trigger on "mlflow", "experiment", "tracking", "model registry", "/experiment-init".Votes: 0GitHub stars: 2
- Eval HarnessFormal evaluation framework implementing eval-driven development (EDD) — define evals (capability + regression), run checks, generate reports, pass@3 metrics. Trigger on "/eval", "evals", "evaluation harness", "pass@3".Votes: 0GitHub stars: 2
- Fullstack WebUse when building Next.js App Router web apps — TypeScript strict, Server Components default, Server Actions for mutations, Tailwind v4, shadcn/ui, Vercel/Netlify deploy. Trigger on "next.js", "nextjs", "app router", "react component", "tailwind", "shadcn".Votes: 0GitHub stars: 2
- Llm EngineeringUse when building LLM-powered applications — structured outputs via Pydantic, prompt versioning, token/cost tracking, evaluation harness integration, retry/caching patterns. Trigger on "structured output", "token count", "prompt version", "llm eval", "pydantic ai", "anthropic sdk".Votes: 0GitHub stars: 2
- Local Ml GpuUse when running ML training/inference on local AMD GPU (RX 6700S / RX 6800S / RX 6650 XT family = Navi 23 RDNA 2). Activates unsloth venv with bundled ROCm + PyTorch, exposes GPU selection pattern, Ollama integration, libmpi workaround for system Python. Trigger on "GPU", "ROCm", "torch.cuda", "ollama", "RX 6700S", "/gpu-check".Votes: 0GitHub stars: 2
- Ml DeploymentUse when deploying ML models to production — model serving frameworks (FastAPI, BentoML, vLLM, Ollama, KServe, Triton), monitoring, autoscaling, ONNX/TensorRT optimization, GPU serving. Trigger on "deploy model", "model serving", "inference", "vllm", "triton", "fastapi model".Votes: 0GitHub stars: 2
- Natural WritingUse when writing documentation (README, docs, blog, commit messages) to avoid AI-tells — banned phrases, active voice, concrete examples, direct start. Trigger on "write readme", "natural writing", "anti-ai", "human voice", "docs style".Votes: 0GitHub stars: 2
- Notebook HygieneUse when working with Jupyter notebooks (.ipynb) — strip output before commit, pin kernels, ensure reproducibility, prevent JSON bloat in git, and integrate with mlflow experiments. Trigger on "notebook", ".ipynb", "jupyter", "strip output", "kernel".Votes: 0GitHub stars: 2
- Project Guidelines ExampleExample project-specific skill template based on a real production app (Zenith) — shows structure for project conventions, commit style, features, infra. Trigger on "project guidelines", "zenith", "project conventions", "template".Votes: 0GitHub stars: 2
- Rag PatternsUse when building RAG (Retrieval-Augmented Generation) systems — chunking strategies, embedding models, vector stores, retrieval (BM25/hybrid), reranking, evaluation, anti-patterns. Trigger on "rag", "retrieval", "embedding", "vector db", "chunking", "/rag-design".Votes: 0GitHub stars: 2
- Remote WorkflowUse when accessing development machine from phone or another device — Tailscale mesh VPN setup, SSH hardening, code-server optional, mosh for roaming, security checklist. Trigger on "remote", "tailscale", "ssh", "from phone", "HP", "/remote-setup".Votes: 0GitHub stars: 2
- Secret ScanUse when committing, pushing, or auditing code — blocks any secret (API key, token, password, private key, .env contents) from reaching GitHub. Combines gitleaks scan + custom grep fallback for known-bad patterns. Trigger on "secret", "leak", ".env", "API key", "token", "/secret-scan", "before push".Votes: 0GitHub stars: 2
- Setup Agent SkillsConfigure this repo for the engineering skills: set up its issue tracker, triage label vocabulary, and domain doc layout. Run once before first use of the other engineering skills.Votes: 0GitHub stars: 2
- Skill RouterAsk which skill or flow fits your situation. A router over the skills in this repo.Votes: 0GitHub stars: 2
- Sql PatternsUse when writing SQL queries — optimization, indexing, CTE, window functions, query plans, warehouse-specific patterns (Postgres, BigQuery, Snowflake, ClickHouse, DuckDB). Trigger on "sql", "query", "select", "join", "explain analyze", "CTE".Votes: 0GitHub stars: 2
- Strategic CompactSuggests manual context compaction at logical intervals to preserve context through task phases rather than arbitrary auto-compaction.Votes: 0GitHub stars: 2
- Tdd WorkflowUse this skill when writing new features, fixing bugs, or refactoring code. Enforces test-driven development with 80%+ coverage including unit, integration, and E2E tests.Votes: 0GitHub stars: 2
- Telegram NotifyUse when sending notifications from long-running scripts (training, deploy, cron) to Telegram phone — setup bot via BotFather, polling atau push, anti-spam, secure token storage. Trigger on "telegram", "notify", "bot", "long-running", "/telegram-setup".Votes: 0GitHub stars: 2
- To SpecTurn the current conversation into a spec and publish it to the project issue tracker: no interview, just synthesis of what you've already discussed.Votes: 0GitHub stars: 2
- To TicketsBreak a plan, spec, or the current conversation into a set of tracer-bullet tickets, each declaring its blocking edges, published to the configured tracker (edges as text in one file per ticket locally, or native blocking links on a real tracker).Votes: 0GitHub stars: 2
- TriageMove issues and external PRs through a state machine of triage roles, categorise, verify, grill if needed, and write agent-ready briefs.Votes: 0GitHub stars: 2
- Verification LoopComprehensive production verification system — build, typecheck, lint, test suite with coverage, security scan, diff review, verdict report. Trigger on "/verify", "verification", "quality gate", "before PR", "before commit".Votes: 0GitHub stars: 2
- WayfinderPlan a huge chunk of work (more than one agent session can hold) as a shared map of decision tickets on your issue tracker, and resolve them one at a time until the way to the destination is clear.Votes: 0GitHub stars: 2