skills.rest
arpitarya
arpitaryaCOMMUNITY
Explore all skills by this author →
WebsiteGitHub

fux-savings

Measures token and dollar cost savings of Fux from real project file sizes.

1|Updated Jun 1, 2026
Search Tags:#token savings#cost tracking#llm cost#roi measurement#documentation index#byte counting#fux
One-click install
npx skills add https://github.com/arpitarya/fux --skill fux-savings-arpitarya
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: fux-savings
Source: https://github.com/arpitarya/fux/tree/main/archive/v0.1/fux/data/skills/savings
Command: npx skills add https://github.com/arpitarya/fux --skill fux-savings-arpitarya

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Teams adopting a documentation index like Fux often cannot prove whether it actually reduces context-window and API costs. This Skill replaces illustrative ROI estimates with measured numbers computed from the project's real file sizes, giving an auditable answer to "is Fux actually saving me anything?" ## Core Features & Use Cases - Measured ROI report: Counts bytes across the index, rules, and governed code files, applying a transparent ~4-chars-per-token heuristic identically to both sides of the comparison. - Dollar pricing: Converts token counts into dollars at a configurable usd_per_mtok rate (defaulting to Claude Opus 4.8's $5/M input rate), with per-lookup and aggregate savings multipliers. - Cumulative tracking: When cost_tracking is enabled, reports lifetime tokens and dollars saved plus per-day, week, and month rates. - Use Case: Run /fux savings "how do we roll back a release" to compare the cost of reading whole governed files versus a Fux-guided lookup, then use a low multiplier as a signal to link more code_refs with fux refs or fux coverage. ## Quick Start Ask the assistant to run /fux savings on a specific question to see the measured token and dollar savings for that lookup.

Frequently Asked Questions about fux-savings

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I measure token savings from a documentation index?▼

Run /fux savings with an optional question after building the Fux index. It counts bytes across the index, rules, and governed code files, applies a 4-chars-per-token heuristic to both sides, and reports per-lookup and aggregate savings multipliers in tokens and dollars.

How to calculate LLM API cost savings in dollars?▼

The savings report prices token counts at a configurable usd_per_mtok rate, defaulting to Claude Opus 4.8's $5 per million input tokens. Every figure in the corpus, per-lookup, and aggregate blocks appears in both tokens and dollars.

Does measuring Fux savings require an LLM or API key?▼

No. The measurement is pure byte-counting with a deterministic heuristic, costing $0 and using no LLM anywhere on the path. It only requires the Fux engine and a freshly built index.

Why is my savings multiplier low or showing a dash?▼

A low or missing multiplier usually means thin code_refs, where rules are not yet linked to the lines they govern, or a very small governed surface. Use fux refs or fux coverage to find undocumented hot files and author rules for them.

Can I track cumulative cost savings over time?▼

Yes. When cost_tracking is set to true, the report adds a Cumulative block showing lifetime tokens and dollars saved plus per-day, per-week, and per-month savings rates.

Related Skills

Explore More Software Engineering Skills→
StanfordSpeziStanfordSpezi
22

project-wiki

Ingest sources into a structured, AI-maintained project wiki with cross-linking.

Official
Advanced
LazyIsEfficientLazyIsEfficient
15

codebase-cost-estimator

Convert measured LOC and architectural complexity into development hours and total cost.

Official
Advanced
brainbrewlabsbrainbrewlabs
7

doc-generation

Generate JSDoc, docstrings, README files, and API references from JavaScript/TypeScript and Python codebases.

Official
Intermediate
ContextJet-aiContextJet-ai
29

estimate-llm-cost

Calculate and project LLM API call costs from token counts.

Official
Basic