مرور مهارت‌ها

This skill should be used when the user asks to "create a hook", "add a PreToolUse/PostToolUse/Stop hook", "validate tool use", "implement prompt-based hooks", "use ${CLAUDE_PLUGIN_ROOT}", "set up event-driven automation", "block dangerous commands", or mentions hook events (PreToolUse, PostToolUse, Stop, SubagentStop, SessionStart, SessionEnd, UserPromptSubmit, PreCompact, Notification). Provides comprehensive guidance for creating and implementing Claude Code plugin hooks with focus on advanced prompt-based hooks API.

OpenAI Codex CLI wrapper — three modes. Code review: independent diff review via codex review with pass/fail gate. Challenge: adversarial mode that tries to break your code. Consult: ask codex anything with session continuity for follow-ups. The "200 IQ autistic developer" second opinion. Use when asked to "codex review", "codex challenge", "ask codex", "second opinion", or "consult codex". (gstack) Voice triggers (speech-to-text aliases): "code x", "code ex", "get another opinion".

Systematically QA test a web application and fix bugs found. Runs QA testing, then iteratively fixes bugs in source code, committing each fix atomically and re-verifying. Use when asked to "qa", "QA", "test this site", "find bugs", "test and fix", or "fix what's broken". Proactively suggest when the user says a feature is ready for testing or asks "does this work?". Three tiers: Quick (critical/high only), Standard (+ medium), Exhaustive (+ cosmetic). Produces before/after health scores, fix evidence, and a ship-readiness summary. For report-only mode, use /qa-only. (gstack) Voice triggers (speech-to-text aliases): "quality check", "test the app", "run QA".

Pre-landing PR review. Analyzes diff against the base branch for SQL safety, LLM trust boundary violations, conditional side effects, and other structural issues. Use when asked to "review this PR", "code review", "pre-landing review", or "check my diff". Proactively suggest when the user is about to merge or land code changes. (gstack)

Comprehensive document creation, editing, and analysis with support for tracked changes, comments, formatting preservation, and text extraction. When Claude needs to work with professional documents (.docx files) for: (1) Creating new documents, (2) Modifying or editing content, (3) Working with tracked changes, (4) Adding comments, or any other document tasks

Parameter-efficient fine-tuning for LLMs using LoRA, QLoRA, and 25+ methods. Use when fine-tuning large models (7B-70B) with limited GPU memory, when you need to train <1% of parameters with minimal accuracy loss, or for multi-adapter serving. HuggingFace's official library integrated with transformers ecosystem.

Generate or edit raster images when the task benefits from AI-created bitmap visuals such as photos, illustrations, textures, sprites, mockups, or transparent-background cutouts. Use when Codex should create a brand-new image, transform an existing image, or derive visual variants from references, and the output should be a bitmap asset rather than repo-native code or vector. Do not use when the task is better handled by editing existing SVG/vector/code-native assets, extending an established icon or logo system, or building the visual directly in HTML/CSS/canvas.

Produce a world-class single-page editorial landing site in the Atelier Zero visual language (Monocle / Apartamento / Études editorial collage) — the same aesthetic Open Design uses for its own marketing surface. The agent fills a typed `inputs.json` from a brand brief, optionally generates 16 collage assets via gpt-image-2, then runs a pure-function composer that emits a self-contained HTML file; a separate path can mirror the Astro marketing site in `apps/landing-page/`. Drop-in scroll-reveal motion and a Headroom-style sticky nav are wired automatically.

Use when the user asks to generate or edit images via the OpenAI Image API (for example: generate image, edit/inpaint/mask, background removal or replacement, transparent background, product shots, concept art, covers, or batch variants); run the bundled CLI (`scripts/image_gen.py`) and require `OPENAI_API_KEY` for live calls.

PostgreSQL-based semantic and hybrid search with pgvector and ParadeDB. Use when implementing vector search, semantic search, hybrid search, or full-text search in PostgreSQL. Covers pgvector setup, indexing (HNSW, IVFFlat), hybrid search (FTS + BM25 + RRF), ParadeDB as Elasticsearch alternative, and re-ranking with Cohere/cross-encoders. Supports vector(1536) and halfvec(3072) types for OpenAI embeddings. Triggers: pgvector, vector search, semantic search, hybrid search, embedding search, PostgreSQL RAG, BM25, RRF, HNSW index, similarity search, ParadeDB, pg_search, reranking, Cohere rerank, pg_trgm, trigram, fuzzy search, LIKE, ILIKE, autocomplete, typo tolerance, fuzzystrmatch

Expert PR and media relations guidance for earned media, press coverage, and reputation building. Use when writing press releases, crafting media pitches, developing journalist relationships, planning embargo strategies, managing crisis communication, placing thought leadership, building analyst relations, submitting award applications, or measuring PR coverage. Use for media outreach, press kits, speaking opportunities, and earned media strategy.

Karpathy's LLM Wiki: build/query interlinked markdown KB.

Build real-time conversational AI voice engines using async worker pipelines, streaming transcription, LLM agents, and TTS synthesis with interrupt handling and multi-provider support

Comprehensive document creation, editing, and analysis with support for tracked changes, comments, formatting preservation, and text extraction. When Claude needs to work with professional documents (.docx files) for: (1) Creating new documents, (2) Modifying or editing content, (3) Working with tracked changes, (4) Adding comments, or any other document tasks

TRL: SFT, DPO, PPO, GRPO, reward modeling for LLM RLHF.

Outlines: structured JSON/regex/Pydantic LLM generation.

DSPy: declarative LM programs, auto-optimize prompts, RAG.

lm-eval-harness: benchmark LLMs (MMLU, GSM8K, etc.).

Build, test, inspect, install, and deploy MCP servers with FastMCP in Python. Use when creating a new MCP server, wrapping an API or database as MCP tools, exposing resources or prompts, or preparing a FastMCP server for Claude Code, Cursor, or HTTP deployment.

Control LLM output with regex and grammars, guarantee valid JSON/XML/code generation, enforce structured formats, and build multi-step workflows with Guidance - Microsoft Research's constrained generation framework