Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx skills add DangerousYams/muxer --skill muxgit clone --depth 1 https://github.com/DangerousYams/muxerWrote 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/dangerousyams/muxer/mux)<a href="https://agentmods.dev/skills/dangerousyams/muxer/mux"><img src="https://agentmods.dev/badge/skills/dangerousyams/muxer/mux.svg" alt="Measured on agentmods" 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 | $0.00042 | $0.01370 |
| Opus 5 | $0.00021 | $0.00685 |
| Sonnet 5 | $0.00008 | $0.00274 |
| Haiku 4.5 | $0.00004 | $0.00137 |
Grade A, and why
mux 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 7d 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.
How it starts
The opening of the file, as written. The whole thing — 63 lines — stays where its author put it; the contents beside it link to each section on GitHub.
muxer — model multiplexer status and doctrine
When invoked, do the following:
- State which model this session's main loop runs on (you know your own model) and what that implies: Fable main loop = premium extra-usage billing, so delegation matters most; Opus or below = Max-plan quota, delegation still saves quota but the stakes are lower.
- Print the routing table and cost table below, then answer the user's question (the skill argument) if one was given.
- Keep it brief and factual. No advice they didn't ask for.
Prime rule
Quality is the constraint, cost is optimized within it. Route each task to the cheapest model that can do it to full quality; when unsure, route up a tier. Substandard delegate output defeats the purpose — rework costs more than the tokens saved. Big meaty tasks are never delegated whole: decompose, brief with acceptance criteria, build one exemplar at a strong tier, fan out the pattern, and verify everything through the reviewer.
Routing table
| Delegate | Model | Use for |
|---|---|---|
| muxer:scout | Haiku | Exploration, file-finding, reading and summarizing logs/docs/code. Read-only. |
| muxer:runner | Haiku | Fire builds, tests, dev servers; condensed pass/fail with the error lines that matter. Never edits code. |
| muxer:writer | Sonnet | Docs, copy, boilerplate, config, mechanical low-risk edits. |
| muxer:pilot | Sonnet | Drive browser/app flows, capture screenshots and GIFs to files. Facts and artifact paths only; taste verdicts go to arbiter. Needs a browser MCP (Claude in Chrome) in the session. |
| muxer:researcher | Sonnet | Web research with named sources. Prefer over general-purpose for research: a cheaper tier (the guard caps general-purpose at Opus) with a tighter sourcing contract. |
| muxer:builder | Opus | Implementing and debugging code. The default for code changes. |
| muxer:reviewer | Opus | Verifying completed work against the original intent. Read-only. |
| muxer:arbiter | Fable (low effort) | Quick top-tier verdicts: taste checks on rendered output, design calls, plan sanity checks. Condensed question in, decision out. Premium billing, but a fraction of oracle's thinking cost. |
| muxer:oracle | Fable | Escalation: visual/creative direction, hardest design decisions. Premium billing — one shot, full context. |
| muxer:codex | external | OpenAI Codex CLI. Work bills to the user's OpenAI account, zero Anthropic tokens. |
| muxer:gemini | external | Google Gemini CLI. Work bills to the user's Google account, zero Anthropic tokens. Huge-context summarization. |
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.
- 7d ago First seen · 63 lines · 42 tokens per session scan A 8b680e9d0aff
mux is a skill published in the GitHub repository DangerousYams/muxer (4 stars, last pushed 24d ago), licensed MIT. It adds 42 tokens to every session and 1,370 once invoked, about $0.0002 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.
Other skills, from other repositories
systematic-debugging
Use when encountering any bug, test failure, or unexpected behavior, before proposing fixes.
local-ai-agents
Build local-first AI agents that run entirely on a developer workstation with Microsoft Foundry Local and Qwen function-calling models. Covers Small Language Models (SLMs), the OpenAI-compatible local endpoint, sandboxed local tools, local RAG with Chroma, local MCP servers, hybrid cloud/local routing, and the…
next-cache-components-adoption
Turn on Cache Components in a Next.js app and resolve the blocking routes it surfaces. Use when the user wants to enable, adopt, or migrate to Cache Components, flip the cacheComponents flag, work through a flood of blocking-prerender / instant validation errors, run the cache-components-instant-false codemod, or…
next-partial-prefetching-adoption
Turn on Partial Prefetching in a Next.js app and work through the insights it surfaces. Use when the user wants to enable or adopt Partial Prefetching, flip the partialPrefetching flag, opt routes in with export const prefetch = 'partial', audit Link prefetch={true} behavior, preserve existing prefetched UI with…
chronicle
Analyze Copilot session history for standup reports, usage tips, session search, and session reindexing. Use when the user asks for a standup, daily summary, usage tips, workflow recommendations, wants to search or find past sessions by keyword/file/PR, wants to reindex their session store, or asks about deleting…
babysit-pr
Babysit a GitHub pull request after creation by continuously polling review comments, CI checks/workflow runs, and mergeability state until the PR is merged/closed or user help is required. Diagnose failures, retry likely flaky failures up to 3 times, auto-fix/push branch-related issues when appropriate, and keep…