local-fusion: Skill for Claude Code

.pi/skills/local-fusion/SKILL.md

local-fusion is a skill for Claude Code, Codex from kelvincushman/local-fusion. It costs 164 tokens per session (2,682 once invoked), scanned A, original, MIT.

A local tool that asks a panel of AI models for answers, has a judge compare them as structured data, and has a synthesizer produce one final answer. It runs through a command-line interface rather than replacing the main AI model.

In plain words
What is it for?
Use it to sanity-check important decisions, compare multiple proposed solutions, and get a deliberated response from local or compatible AI models.
Why use it?
It provides a second opinion when a plan, design, answer, or code decision needs more than one perspective. The returned trace shows where the models agreed, disagreed, or missed something.

Skill for Claude CodeCodex

Written for no agent in particular: nothing here depends on one. Also seen: mentions Claude Code; mentions Codex.

This is kelvincushman/local-fusion's own configuration. It tells Claude Code and Codex how to work on local-fusion itself, so it is not a mod to install elsewhere. Copy it as a starting point and replace the rules that are about this project. Everything local-fusion configures →

Not installable: its command points at a path on the author’s own machine, so it runs nowhere else. The line is /Users/kelvincushman/dev/local-fusion.

Reuse

Borrowing it

Nothing to install: this file belongs to kelvincushman/local-fusion. 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/kelvincushman/local-fusion/main/.pi/skills/local-fusion/SKILL.md
Clone the repo
git clone --depth 1 https://github.com/kelvincushman/local-fusion

Made for: Claude Code, Codex.

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.

agentmods badge for local-fusion

README.md
[![agentmods](https://agentmods.dev/badge/skills/kelvincushman/local-fusion/local-fusion/github.svg)](https://agentmods.dev/skills/kelvincushman/local-fusion/local-fusion)
Your own site
<a href="https://agentmods.dev/skills/kelvincushman/local-fusion/local-fusion"><img src="https://agentmods.dev/badge/skills/kelvincushman/local-fusion/local-fusion/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.

agentmods 80×15 button for local-fusion

Your own site · 80×15
<a href="https://agentmods.dev/skills/kelvincushman/local-fusion/local-fusion"><img src="https://agentmods.dev/badge/skills/kelvincushman/local-fusion/local-fusion.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 164 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 2,682 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.
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.

ModelPer sessionOnce invoked
Fable 5.1 $0.00164 $0.02682
Opus 5 $0.00082 $0.01341
Sonnet 5 $0.00033 $0.00536
Haiku 4.5 $0.00016 $0.00268

Measured 12d ago against content hash 50eb01df44c8, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-12, from the pricing page.

Security

Grade A, and why

local-fusion 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 12d 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.

.pi/skills/local-fusion/SKILL.md · 182 lines

How it starts

The opening of the file, as written. The whole thing — 182 lines — stays where its author put it; the contents beside it link to each section on GitHub.

local-fusion — local multi-model council (as a tool)

local-fusion sends one prompt to a panel of local models, has a judge compare their answers as structured JSON, then a synthesizer writes a single final answer. You invoke it through bash. It returns a synthesized answer plus a structured trace of where the models agreed, disagreed, and what they missed.

This is a tool you call, not your model backend. Your own model still drives this session — reach for local-fusion when a deliberated second opinion is worth ~20–40s.

Invoke it with the built-in bash tool. There is NO function or tool named local-fusion — calling one will fail with "Tool local-fusion not found". To use the council you run the CLI below via bash, e.g. bash → node src/cli.mjs ask "<question>", and read the answer from the command's stdout.

Repo root: /Users/kelvincushman/dev/local-fusion — run the commands below from there.

Preconditions — check before real use

The shipped config drives models through a Pi subprocess (backend: "pi"), so the only precondition is that Pi itself is installed and authenticated for the models in local-fusion.config.json:

# 1. Pi is installed and on PATH.
command -v pi >/dev/null && pi --version || echo "pi not installed"

# 2. Pi is authed for the providers used in the config. This lists every model Pi
#    can currently reach (looks for glm-5.2 / gpt-5.4 / claude-opus-4-8):
pi --list-models 2>/dev/null | grep -Ei 'glm-5.2|gpt-5.4|claude-opus-4-8' || echo "a target model is not authed — run: pi /login"

If a precondition fails, tell the user how to fix it (install Pi, run pi /login, or pick a different provider in the config). Don't invent auth silently.

Opus / Claude Pro-Max caveat: Anthropic subscription access through a third-party harness (Pi) is billed per-token from your extra usage budget, not your plan limits. If Opus errors with 400 ... Add more at claude.ai/settings/usage, the account has no/used-up extra-usage budget — the user must add it at claude.ai/settings/usage. GLM-5.2 (free) and GPT-5.4 (ChatGPT sub) do not have this restriction.

Read the full file on GitHub · 182 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.

  1. 12d ago First seen · 182 lines · 164 tokens per session scan A 50eb01df44c8

Subscribe to this mod's changes

local-fusion is a skill published in the GitHub repository kelvincushman/local-fusion (0 stars, last pushed 2mo ago), licensed MIT. It adds 164 tokens to every session and 2,682 once invoked, about $0.0008 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 2026-08-31.

Related

Other skills, from other repositories

browser-use

Direct browser control via CDP for web interaction: automation, scraping, testing, screenshots, and site/app work.

browser-use/browser-use · 26 tokens

open-source

Documentation reference for writing Python code using the browser-use open-source library. Use this skill whenever the user needs help with Agent, Browser, or Tools configuration, is writing code that imports from browseruse, asks about @sandbox deployment, supported LLM models, Actor API, custom tools, lifecycle…

browser-use/browser-use · 137 tokens

ocx

Drive a running opencodex (ocx) proxy from the CLI — account pools, provider routing, model catalog, usage and cost attribution, request logs, access keys, storage cleanup, and the management API. Use when a task involves controlling or inspecting an opencodex proxy rather than editing the opencodex codebase.…

lidge-jun/opencodex · 119 tokens

crit-story

Author a crit story and continue the interactive review loop only when the user explicitly invokes crit-story or directly asks you to generate a crit story. Do not infer this skill from generic review, PR, or diff-review requests.

tomasz-tomczyk/crit · 47 tokens

snip

You are an expert at writing declarative YAML filters for snip, a CLI proxy that reduces LLM token consumption by filtering shell output.

edouard-claude/snip · 0 tokens

goai

GoAI is a Go SDK for AI applications. One unified API across 25+ LLM providers. Inspired by the Vercel AI SDK, adapted to Go idioms (generics, interfaces, channels).

zendev-sh/goai · 0 tokens