Name: local-llm-expert
Rating: 59.2 (1 reviews)
Author: administrakt0r

local-llm-expert

Master local LLM inference, model selection, VRAM optimization, and local deployment using Ollama, llama.cpp, vLLM, and LM Studio. Expert in quantization formats (GGUF, EXL2) and local AI privacy.

Quality

68%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Passed

No findings from the security scan

Fix and improve this skill with Tessl

tessl review fix ./plugins/AI-Agents-Safe-Coding-Skills-claude/skills/local-llm-expert/SKILL.md

The canonical home for this skill is local-llm-expert in administrakt0r/AI-Agents-Safe-Coding-Skills

Quality

Content

67%Weight 40%Scale 1-5

Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.

The body is well-structured and concrete, with real commands, formulas, and templates, but carries redundant restatements and prose-described examples. Tightening redundancy and showing actual example commands would lift conciseness and actionability.

Suggestions

Remove the 'Purpose' section (it restates the frontmatter description) and trim 'Knowledge Base' items that restate well-known facts like 'Complete catalog of GGUF formats' to improve conciseness.

Replace the prose-described 'Example Interactions' with at least one fully shown executable command/Modelfile snippet to move actionability toward 5.

Dimension	Reasoning	Score
Conciseness	The body is mostly efficient and well-sectioned, but the 'Purpose' section restates the frontmatter and 'Knowledge Base' lists concepts Claude already knows ('Complete catalog of GGUF formats', 'Benchmarks for Llama 3'). Matches the 'mostly efficient but some unnecessary explanation' anchor; not a 2 because structure keeps padding bounded.	3 / 5
Actionability	Provides concrete, executable-level specifics (VRAM formula 'Parameters * Bits-per-weight / 8', flags -ngl/-c/-m, a real ChatML string '<\|im_start\|>system\n...'), but the 'Example Interactions' describe outputs in prose rather than giving copy-paste commands/code. Matches 'mostly executable guidance; minor gaps'.	4 / 5
Workflow Clarity	Both 'Instructions' and 'Response Approach' give clear 5-step sequences. No destructive/batch operations, so the validation cap does not apply; absent explicit validation checkpoints keeps it below 5. Matches 'clear sequence with most checkpoints present'.	4 / 5
Progressive Disclosure	No bundle files exist; the single self-contained file is well-organized with clear headers. Slightly over the 50-line simple-skill threshold and some 'Knowledge Base' content is inlined rather than externalized, matching 'good structure; minor organization gaps'.	4 / 5
	Total	15 / 20 Passed

Description

70%Weight 40%Scale 1-5

Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.

The description is specific, trigger-rich, and clearly distinct, but lacks an explicit 'Use when...' clause, which caps its completeness. Adding trigger guidance would raise the overall quality.

Suggestions

Append an explicit 'Use when...' clause naming natural triggers (e.g., 'Use when running models locally with Ollama/llama.cpp, choosing quantization formats, or sizing VRAM for offline LLM deployment').

Add common synonyms like 'self-hosted LLM', 'offline AI', or 'run models locally' to broaden trigger-term coverage.

Dimension	Reasoning	Score
Specificity	Lists several concrete action areas ("local LLM inference, model selection, VRAM optimization, and local deployment") plus named tools and formats, matching the 'several specific actions; minor gaps' anchor. Not a 5 because coverage of actions is not exhaustive.	4 / 5
Completeness	The 'what' is clearly stated but there is no 'Use when...' clause or equivalent explicit trigger guidance, which per the rubric caps completeness at 3.	3 / 5
Trigger Term Quality	Includes natural terms users actually say for local LLM work (Ollama, llama.cpp, vLLM, LM Studio, GGUF, EXL2, VRAM, quantization), giving good keyword coverage. Not a 5 because common synonyms like 'run models locally', 'offline LLM', or 'self-hosted' are missing.	4 / 5
Distinctiveness Conflict Risk	Occupies a clear niche (local/offline LLM deployment) with highly specific tool and format triggers (Ollama, llama.cpp, GGUF, EXL2) that minimize conflict with other skills.	5 / 5
	Total	16 / 20 Passed

Validation

93%

Warnings & errors only

Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.

Validation — 15 / 16 Passed

Validation for skill structure

Criteria	Description	Result
frontmatter_unknown_keys	Unknown frontmatter key(s) found; consider removing or moving to metadata	Warning

	Total	15 / 16 Passed

Repository: administrakt0r/AI-Agents-Safe-Coding-Skills
Path: plugins/AI-Agents-Safe-Coding-Skills-claude/skills/local-llm-expert/SKILL.md
Commit: 57c135b

Reviewed: 4 days ago

Table of Contents

Discovery Implementation Validation

Is this your skill?

If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.