llamacpp skills

30 tagged llamacpp, measured the same way as everything else here.

Browse within: local-ai 22GGUF 14artificial-intelligence 14box 14litert 14exllama 8hosting 8local 8

gptme/gptme

Skill Claude CodeCodex

Classify codebases before modification to choose appropriate development approach.

4.4k 2d ago A 16 tokens original MIT

python-repl

02

gptme/gptme

Skill Claude CodeCodex

Interactive Python REPL automation with common helpers and best practices.

4.4k 2d ago A 15 tokens original MIT

automations

03

sybil-solutions/local-studio

Skill Claude CodeCodex

Manage Local Studio automations — scheduled prompts the app re-runs on its own — when the user asks to schedule, pause, resume, edit, inspect or delete recurring work, or asks what a scheduled job has been doing.

1.7k 5d ago A 49 tokens original Apache-2.0

github

04

sybil-solutions/local-studio

Skill Claude CodeCodex

GitHub through the gh CLI that is already installed and signed in on this machine — issues, pull requests, diffs, reviews, Actions runs, code search, and the REST API. Load when the user asks about a repo, an issue, a PR, a review, red CI, or anything on github.com.

1.7k 5d ago A 67 tokens original Apache-2.0

obsidian

05

sybil-solutions/local-studio

Skill Claude CodeCodex

The user's Obsidian vault — searching, reading, creating and appending to their markdown notes, with wikilinks, frontmatter and tags handled the way Obsidian means them. Load when the user mentions their notes, their vault, Obsidian, a daily note, a note by name, or asks you to write something down for them.

1.7k 5d ago A 75 tokens original Apache-2.0

running-all-tests

06

intentee/paddler

Skill Claude CodeCodex

Runs every test suite in the paddler workspace on the fastest available device. Use when the user asks to run the tests, run all the tests, run the full test suite, or check that everything still passes.

1.7k 1mo ago A 47 tokens original Apache-2.0

running-coverage

07

intentee/paddler

Skill Claude CodeCodex

Runs every test suite in the paddler workspace on the fastest available device, and produces code coverage report. Use when the user asks to run the code coverage, or to check the coverage.

1.7k 1mo ago A 42 tokens original Apache-2.0

mood-tracker

08

jegly/Box

Skill Claude CodeCodex

A simple mood tracking skill that stores your daily mood and comments. Use this when the user wants to log their mood, track how they feel, or see their mood history.

790 8d ago A 40 tokens

translator

09

jegly/Box

Skill Claude CodeCodex

Translates text between languages. Use this when the user wants to translate a word, phrase, or passage into another language.

790 8d ago A 27 tokens

kitchen-adventure

10

jegly/Box

Skill Claude CodeCodex

Act as a dungeon master for a text-based adventure set in a world where everyone is a sentient kitchen appliance. Trigger when user says "start kitchen adventure".

790 8d ago A 37 tokens

local-offload-setup

11

dmmdea/offload-harness

Skill Claude CodeCodex

Use when setting up the "local-offload" harness on a Windows machine — a free local Gemma-4 cascade that lets a coding agent delegate short-context grunt work (summarize / classify / extract / triage) so those tokens never hit the cloud context. Cross-vendor: NVIDIA (CUDA, ≥8GB), AMD Radeon incl. RDNA3 iGPUs like the…

2 2d ago A 204 tokens original Apache-2.0

pp-comfyui

12

dmmdea/offload-harness

Skill Claude CodeCodex

Drive a local ComfyUI render server from the shell, with a durable record of every run the server itself forgets. Trigger phrases: submit a comfyui graph, why can't comfyui see my model, how long did that render take, what made this output file, check what this loader accepts, use comfyui, run comfyui.

2 2d ago A 81 tokens original Apache-2.0

pp-llamaswap

13

dmmdea/offload-harness

Skill Claude CodeCodex

The llama-swap operations console — durable history, drain-aware control, and measurement commands with three specific guarantees: keep-set unloads are refused statically by id AND alias (never from server ttl), --drain fails closed when slot state is unreadable, and fit/ctx refuse to answer inside their uncertainty…

2 2d ago A 138 tokens original Apache-2.0

local-fusion

14

kelvincushman/local-fusion

Skill Claude CodeCodex

Use when you want a multi-model "council" second opinion before committing to an answer, plan, design, or code direction. Trigger when the user says "ask the council", "fuse this", "fusion", "second opinion", "get multiple perspectives", "have the panel weigh in", "sanity-check this", "what would other models say", or…

0 2mo ago A 164 tokens original MIT