token-cost

Classical formal logic from G. Chelpanov's textbook: concepts, judgments, syllogisms, induction, fallacies, laws of thought.

1 skill · 1 command

Cross-review code with several console code agents at once: Claude Code, Codex, Coddy, Cursor Agent, Devin, OpenCode and others review the same brief in parallel and blind to each other, then the orchestrating agent verifies every finding against the code and decides alone what matters and what to fix. Linux, macOS and Windows; reviewers chosen once per agent, for one project or for all of them.

1 skill

Use when the user asks to create, draw, render, or update a technical diagram, architecture scheme, process flow, lifecycle, or system map from supplied information and expects editable source plus a finished image.

1 skill

Prepare a Linux server with NVIDIA or Intel Arc GPUs for LLM and neural-network workloads: NVIDIA driver + CUDA from the official repo or the in-tree Intel i915/xe driver with distro firmware, Docker Engine with verified GPU access (NVIDIA Container Toolkit or /dev/dri passthrough) as the mandatory base, staged verification gates for both vendors, optional docker-compose presets for vLLM, Infinity, OpenWebUI and Ollama plus Intel Arc stacks (llama.cpp SYCL/Vulkan, Ollama, OpenVINO, vLLM), GPU power limits persisted via systemd, case fans driven by GPU temperature and Intel Arc idle-power fixes (ASPM).

1 skill

Generate secure MikroTik RouterOS v7 configs from plain language, or review and harden an existing /export. Hardening baseline, four presets, per-model device reference, offline validator.

1 skill

Search the web, read the clean text of a page by URL, crawl a few pages of one site and search Telegram channels through the NeuralDeep Search API with an sk- key. A standard-library Python client takes the key from --api-key, NEURALDEEP_SEARCH_API_KEY or NEURALDEEP_API_KEY, prints compact text or the raw JSON and exits with meaningful codes.

1 skill

Claude API and SDK documentation skill for building LLM-powered applications

19 skills

Access thousands of AI prompts and skills directly in your AI coding assistant. Search prompts, discover skills, save your own, and improve prompts with AI.

2 skills · 2 commands · 2 agents · 1 MCP server

Security review for Claude-generated code. Pattern-based warnings on edits, LLM-powered diff review on Stop, and an agentic commit reviewer that catches injection, XSS, SSRF, hardcoded secrets, and 25+ other vulnerability classes.

hooks

Token usage tracking, model cost attribution per agent, budget alerts, and optimization recommendations

24 skills · 1 command · 1 agent · hooks

Skill: add headroom-ai compress() to a Python or TypeScript app to cut LLM input tokens (tool outputs, logs, JSON, chat history). No API key.

1 skill

Install the "AAS AI Product & Evaluation Ops" editorial skill bundle for Claude Code.

13 skills