llama-cpp

55 / 100 · Review before installing

Check before installing. Some of what it can do is expected: a coding skill usually runs build and test commands.

By maystudios/claude-skills (GitHub) · coding · 29 stars · scanned at commit 25145cf

Guide for llama.cpp, the C/C++ LLM inference framework by ggml-org. Covers the C API (llama.h), GGUF format, quantization (Q4_K_M, Q8_0, IQ4_XS), CMake builds, GPU backends (CUDA, Vulkan, Metal, ROCm), HTTP server with OpenAI-compatible API, embeddings, grammar constraints, function calling, LoRA, speculative decoding, multimodal, and UE5 integration. Use when: llama.cpp, GGUF models, local LLM inference, llama.h, llama-server, quantizing, ggml, building/linking llama.cpp, GPU acceleration, lla…

Works with any agent that supports SKILL.md skills.

What this skill can do

Downloads and installs other software

The author does not mention this.

Found in 7 places.

llama-cpp/SKILL.md, line 49brew install llama.cpp # macOS/Linux
llama-cpp/references/build-and-integration.md, line 640brew install llama.cpp
llama-cpp/references/build-and-integration.md, line 648brew install llama.cpp

Show 4 more

llama-cpp/references/quantization-guide.md, line 341pip install -r requirements.txt
llama-cpp/references/quantization-guide.md, line 358pip install huggingface-cli
llama-cpp/references/server-api.md, line 1901pip install openai
llama-cpp/references/server-api.md, line 2113pip install anthropic

Runs commands on your computer

The author does not mention this. Expected for a coding skill.

Found in 3 places.

llama-cpp/SKILL.md, line 49brew install llama.cpp # macOS/Linux
llama-cpp/SKILL.md, line 83git clone https://github.com/ggml-org/llama.cpp && cd llama.cpp
llama-cpp/SKILL.md, line 133llama-quantize model-f16.gguf model-q4km.gguf Q4_K_M

Sends data to these websites: huggingface.co

The author does not mention this.

Found in 74 places.

llama-cpp/SKILL.md, line 56curl http://localhost:8080/v1/chat/completions \
llama-cpp/references/quantization-guide.md, line 386wget https://huggingface.co/bartowski/Llama-3.1-8B-Instruct-GGUF/resolve/main/Llama-3.1-8B-Instruct-Q4_K_M.gguf
llama-cpp/references/quantization-guide.md, line 389curl -L -o model.gguf https://huggingface.co/bartowski/Llama-3.1-8B-Instruct-GGUF/resolve/main/Llama-3.1-8B-Instruct-Q4_K_M.gguf

Show 17 more

llama-cpp/references/server-api.md, line 76curl http://localhost:8080/health
llama-cpp/references/server-api.md, line 173curl http://localhost:8080/v1/chat/completions \
llama-cpp/references/server-api.md, line 220curl http://localhost:8080/v1/chat/completions \
llama-cpp/references/server-api.md, line 249curl http://localhost:8080/v1/chat/completions \
llama-cpp/references/server-api.md, line 279curl http://localhost:8080/v1/chat/completions \
llama-cpp/references/server-api.md, line 334curl http://localhost:8080/v1/completions \
llama-cpp/references/server-api.md, line 371curl http://localhost:8080/v1/completions \
llama-cpp/references/server-api.md, line 384curl http://localhost:8080/v1/completions \
llama-cpp/references/server-api.md, line 400curl http://localhost:8080/v1/embeddings \
llama-cpp/references/server-api.md, line 431curl http://localhost:8080/v1/embeddings \
llama-cpp/references/server-api.md, line 450curl http://localhost:8080/v1/models
llama-cpp/references/server-api.md, line 484curl http://localhost:8080/v1/messages \
llama-cpp/references/server-api.md, line 521curl http://localhost:8080/v1/messages \
llama-cpp/references/server-api.md, line 536curl http://localhost:8080/v1/messages \
llama-cpp/references/server-api.md, line 555curl http://localhost:8080/v1/messages/count_tokens \
llama-cpp/references/server-api.md, line 580curl http://localhost:8080/v1/responses \
llama-cpp/references/server-api.md, line 620curl http://localhost:8080/v1/responses \

and 54 more places

Changes files outside its own folder

The author does not mention this.

Found in 1 place.

llama-cpp/references/build-and-integration.md, line 73cmake --install build --prefix /usr/local/llama-cpp

Can call paid AI services billed to you

The author does not mention this.

Found in 8 places.

llama-cpp/references/server-api.md, line 1907from openai import OpenAI
llama-cpp/references/server-api.md, line 1930from openai import OpenAI
llama-cpp/references/server-api.md, line 1951from openai import OpenAI

Show 5 more

llama-cpp/references/server-api.md, line 1996from openai import OpenAI
llama-cpp/references/server-api.md, line 2060from openai import OpenAI
llama-cpp/references/server-api.md, line 2077from openai import OpenAI
llama-cpp/references/server-api.md, line 2117from anthropic import Anthropic
llama-cpp/references/server-api.md, line 2194from openai import AsyncOpenAI

Safer alternatives

Low-risk skills in the same category (coding) that look related. Check that one does what you need.

Install

Paste this into your terminal. It downloads exactly the version we scanned, not whatever is newest.

if [ -e ~/.agents/skills/'llama-cpp' ] || [ -L ~/.agents/skills/'llama-cpp' ]; then echo 'A skill folder named llama-cpp already exists — not replacing it' >&2; false; else d=$(mktemp -d) && git clone --quiet --filter=blob:none --no-checkout 'https://github.com/maystudios/claude-skills' "$d" && git -C "$d" checkout --quiet '25145cf85b0709dcc2f7a40a7035c7527c137a6c' -- 'llama-cpp' && mkdir -p ~/.agents/skills && cp -R "$d"/'llama-cpp' ~/.agents/skills/'llama-cpp' && rm -rf "$d"; fi

Codex reads ~/.agents/skills directly. For Claude Code, Cursor, Copilot, Gemini CLI, Antigravity or OpenCode, also link it. For Claude Code: ln -s ~/.agents/skills/'llama-cpp' ~/.claude/skills/'llama-cpp' (other agents: ~/.cursor/skills, ~/.copilot/skills, ~/.gemini/skills, ~/.gemini/config/skills, ~/.config/opencode/skills). Or install through the finder skill, which does both.

Badge for your README

The badge for maystudios/claude-skills shows the score of its lowest-scoring skill.

SkillTrust badge for maystudios/claude-skills

[![SkillTrust](https://zalavijaysinh.github.io/skilltrust/badges/maystudios/claude-skills.svg)](https://zalavijaysinh.github.io/skilltrust/repos/maystudios/claude-skills/)

Score history

Date	Version	Score
2026-11-07	`25145cf`	55