Borrowing it
Nothing to install: this file belongs to superposed-labs/fluxion-bus. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.
curl -O https://raw.githubusercontent.com/superposed-labs/fluxion-bus/main/.agents/skills/update-model-prices/SKILL.mdgit clone --depth 1 https://github.com/superposed-labs/fluxion-busWrote 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.
[](https://agentmods.dev/skills/superposed-labs/fluxion-bus/update-model-prices)<a href="https://agentmods.dev/skills/superposed-labs/fluxion-bus/update-model-prices"><img src="https://agentmods.dev/badge/skills/superposed-labs/fluxion-bus/update-model-prices/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.
<a href="https://agentmods.dev/skills/superposed-labs/fluxion-bus/update-model-prices"><img src="https://agentmods.dev/badge/skills/superposed-labs/fluxion-bus/update-model-prices.svg" alt="Reviewed on agentmods" width="80" height="20"></a>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 | $0.00172 | $0.03269 |
| Opus 5 | $0.00086 | $0.01635 |
| Sonnet 5 | $0.00034 | $0.00654 |
| Haiku 4.5 | $0.00017 | $0.00327 |
Grade A, and why
update-model-prices scanned grade A with 1 finding 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 10d 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.
Makes network callslowCapability
Not a fault in itself. Listed so you know the mod talks to something, and to what.
Google serve prices in plain HTML, so `curl`/WebFetch is enough for those. How it starts
The opening of the file, as written. The whole thing — 244 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Update model prices
Fluxion estimates the dollar cost of token usage from a hand-maintained price table. There is no official pricing API, and the CLIs run models newer than any public price list, so prices are refreshed by a person running this procedure on demand (a few times a year, when prices change) — not by a scheduled scraper.
Your job here is the judgment-heavy part a cron can't do: render the pages, read the numbers semantically, decide how each model maps into the table, and hand the human a clear, correct proposal. The mechanical fetching/auditing is already scripted.
Where the table lives — edit upstream first
The source of truth is the standalone repo
superposed-labs/llm-price-table
(its SCHEMA.md defines the format). src/fluxion/usage/model_prices.json is a
bundled snapshot of that file — the offline fallback and release baseline.
At runtime price_data.load_price_json picks whichever of the snapshot and the
locally refreshed cache carries the newer top-level updated_at, so the daily
background refresh normally delivers a new price within a day even with no
Fluxion change at all.
A price change therefore lands in two repos, in this order:
- Edit
model_prices.jsonin the llm-price-table checkout; bump itsupdated_at. - Copy that file verbatim over
src/fluxion/usage/model_prices.json.
Keep the snapshot a byte-for-byte copy and diff the two afterwards. Drift is a
silent bug: a family present upstream but missing from the snapshot mispriced
every id that resolved through it (this actually happened with mythos, which
fell through to the coarse claude fallback at half the real rate).
The one hard rule
Propose, never auto-apply. A wrong number silently corrupts every cost
figure in the dashboard, so a human must verify before anything lands. Produce a
clear summary of proposed changes and stop. Only edit the file if the user
explicitly approves the specific numbers, and never git commit — committing
is always the user's call. Git history is the price-change log, so it must stay
human-authored.
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.
- 10d ago First seen · 244 lines · 172 tokens per session scan A 9709cf6b8444
update-model-prices is a skill published in the GitHub repository superposed-labs/fluxion-bus (2 stars, last pushed 3d ago), licensed Apache-2.0. It adds 172 tokens to every session and 3,269 once invoked, about $0.0009 per session on Opus 5. A static security scan graded it A with 1 finding (makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other skills, from other repositories
tmux
Remote-control tmux sessions for interactive CLIs by sending keystrokes and scraping pane output.
summarize
Summarize or extract text/transcripts from URLs, podcasts, and local files (great fallback for “transcribe this YouTube/video”).
cw-gates
Use before claiming any Codewhale change is done, green, or ready to land: the focused-to-broad verification ladder, the budget checks CI enforces, and the rules for what counts as a passing test.
feishu
Work with Feishu or Lark bots, docs, sheets, bitables, approval flows, and OpenAPI/MCP setup without hardcoding credentials.
interview
Ask one useful structured question at a time only when material product/implementation choices are genuinely missing; remember answers and produce a brief/spec. Discoverable facts should be investigated instead of asked.
ppt-review
Read this skill only after current officereview structural output and relevant rendered images exist. Do not use source code, a tool success flag, or the first-slide preview as a substitute for deck evidence.