What is local-inference?

A way to run language models directly on Apple Silicon Macs through Crystal and MLX, Apple's machine-learning framework. Models can support chat, streaming text, and structured JSON output without a cloud API.

Use it for local chat, streamed responses, structured data generation, or on-device model testing on an M1-or-newer Mac.

It keeps inference on the device, avoiding a required connection to services such as OpenAI. The model can remain loaded in memory for reuse.

How crimson-knight/llamero configures Claude Code

Reuse

Borrowing it

Nothing to install: this file belongs to crimson-knight/llamero. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.

Copy the file

curl -O https://raw.githubusercontent.com/crimson-knight/llamero/main/.claude/skills/local-inference/SKILL.md

Clone the repo

git clone --depth 1 https://github.com/crimson-knight/llamero

Made for: Claude Code.

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

README.md

[![agentmods](https://agentmods.dev/badge/skills/crimson-knight/llamero/local-inference/github.svg)](https://agentmods.dev/skills/crimson-knight/llamero/local-inference)

Your own site

<a href="https://agentmods.dev/skills/crimson-knight/llamero/local-inference"><img src="https://agentmods.dev/badge/skills/crimson-knight/llamero/local-inference/github.svg" alt="Measured on agentmods" height="20"></a>

Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.

Your own site · 80×15

<a href="https://agentmods.dev/skills/crimson-knight/llamero/local-inference"><img src="https://agentmods.dev/badge/skills/crimson-knight/llamero/local-inference.svg" alt="Reviewed on agentmods" width="80" height="20"></a>

Per session 86 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.

When invoked 2,862 The whole file, excluding the scripts and references it only reads on demand.

Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe. Third-party audits

NVIDIA SkillSpector pass 7 Sept 2026

How audits are shown

Origin original No closer match found in the catalogue.

Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

Model	Per session	Once invoked
Fable 5.1	$0.00086	$0.02862
Opus 5	$0.00043	$0.01431
Sonnet 5	$0.00017	$0.00572
Haiku 4.5	$0.00009	$0.00286

Measured 11d ago against content hash f93eb59352c4, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-10, from the pricing page.

Security

Grade A, and why

local-inference scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 11d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

.claude/skills/local-inference/SKILL.md · 250 lines

How it starts

The opening of the file, as written. The whole thing — 250 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Local inference with llamero (Apple Silicon / MLX)

llamero runs language models locally on Apple Silicon through a Swift MLX bridge. The model loads into memory once and stays resident; every chat, stream, or structured call reuses it. No Python, no llama.cpp, no cloud API.

Requirements — check these first

macOS on Apple Silicon (M1 or newer). On any other platform, Llamero::Native still compiles and runs, but uses a mock bridge that returns canned text. Check at runtime with runtime.real_bridge?.
The Swift bridge dylib must be built once per machine:
```
# In a project that depends on llamero:
cd lib/llamero/native/llamero-mlx && ./build.sh
# Inside the llamero repo itself:
cd native/llamero-mlx && ./build.sh
```
build.sh installs libLlameroMLXBridge.dylib and mlx.metallib to $LLAMERO_HOME/lib when set, otherwise ~/.llamero/lib/, where llamero finds them from any project. It requires Xcode with the Metal toolchain (xcodebuild -downloadComponent MetalToolchain if the build complains about a missing metal tool).
Models download automatically from Hugging Face on first load into the configured storage root's models/ directory. The mlx-community conversions (including Gemma) are generally ungated and need no token. Gated repos (e.g. google/* originals) need HF_TOKEN set in the environment.

Storage root

Default storage is ~/.llamero. Apps that need app-owned AI data set this at boot before creating runtimes:

Llamero.storage_root = Path.home.join(".scribe")

LLAMERO_HOME=/path/to/root is the env alternative; programmatic wins. The root controls local model downloads (models/), adapter artifacts (adapters/), bridge lookup (lib/), and audio model caches (audio_models/) when audio is configured. For app-owned bridge installs, set LLAMERO_HOME when running build.sh. FluidAudio 0.15.2 still pins English Kokoro G2P assets to its own TTS cache; see development_docs/storage_configuration.md.

Read the full file on GitHub · 250 lines

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

11d ago First seen · 250 lines · 86 tokens per session scan A f93eb59352c4

Subscribe to this mod's changes

local-inference is a skill published in the GitHub repository crimson-knight/llamero (22 stars, last pushed 18d ago), licensed MIT. It adds 86 tokens to every session and 2,862 once invoked, about $0.0004 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.

Data and AI

agent-platform-rag-engine-management

Manage and query Agent Platform RAG Engine Corpora and retrieve grounded contexts using the Google GenAI SDK. Use when listing RAG corpora or files, inspecting a corpus, retrieving contexts, or generating content grounded in a RAG corpus. Do not use for standard database queries (use SQL/Spanner skills), Google…

google/skills · 85 tokens

agent-platform-model-registry

Agent Platform Model Registry Management. Use when you need to upload, list, describe, update, or delete machine learning models (and their versions) in the Agent Platform Model Registry. Don't use for model training, model deployment to endpoints, or managing non-Agent Platform models.

google/skills · 60 tokens

foundry-config-setup

Resolve missing setup caused by a hardcoded Foundry project endpoint or model in a sample. Use when a sample fails because it uses a placeholder/hardcoded projectendpoint (for example "https://your-project.services.ai.azure.com") or a hardcoded model instead of reading them from the environment.

microsoft/agent-framework · 65 tokens

google-cloud-solution-agentic-analytics-spark-knowledge-catalog

Discovers requirements and generates guidance to design and deploy a governed, secure agentic-analytics solution for data that's distributed across Google Cloud, other cloud providers, or on-premises. Data that's outside Google Cloud (such as data from Databricks, Snowflake, Salesforce, SAP, or Oracle systems) is…

google/skills · 138 tokens

training-check

Interactively monitor training metrics from the current Codex session, periodically checking WandB or fallback logs for NaN, divergence, plateaus, and broken runs.

wanshuiyin/Auto-claude-code-research-in-sleep · 35 tokens

nemo-automodel-launcher-config

Configure NeMo AutoModel job launches for interactive runs, Slurm clusters, and SkyPilot cloud execution.

NVIDIA/skills · 30 tokens

llamero: Skill for Claude Code