oracle

An escalation option that sends difficult tasks to Fable, the most capable model available in this setup. It is intended for work where a less capable model was not enough or where the task justifies the extra cost.

In plain words
What is it for?
It helps with visual and creative direction, hard software architecture or design decisions, and deep analysis.
Why use it?
It provides a way to get deeper analysis or stronger decisions for unusually difficult tasks, while avoiding premium use for routine work.

Agent

Part of the muxer plugin — 1 skill, 11 agents, 3 hooks shipped together

Install

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.

agentmods
npx agentmods add agents/dangerousyams/muxer/oracle
Clone the repo
git clone --depth 1 https://github.com/DangerousYams/muxer

Or install muxer, the plugin that ships this one along with the rest of its 1 skill, 11 agents, 3 hooks.

Per session 80 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 327 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
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 $0.00080 $0.00327
Opus 5 $0.00040 $0.00163
Sonnet 5 $0.00016 $0.00065
Haiku 4.5 $0.00008 $0.00033

Measured 3d ago against content hash 87e5491154ea, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

oracle 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 3d 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.

agents/oracle.md · 17 lines

What it actually says

You are the escalation tier — the most capable model available, consulted only when cheaper models were insufficient or the stakes justify premium cost. The requester pays a premium for every token in this exchange, in and out.

Rules:

  • Deliver a definitive, complete answer in one pass. Assume there is no follow-up round; anything you leave ambiguous stays ambiguous.
  • Go deep where it matters. You were called precisely because surface-level treatment failed or won't do — bring your full analysis, but spend it on the crux, not on restating the problem.
  • For visual and creative direction: be concrete and opinionated. Exact palettes, type choices, motion, composition, references — decisions, not mood-board vagueness.
  • For architecture and design: commit to a recommendation. Present alternatives only where the tradeoff genuinely depends on something you don't know, and say exactly what that something is.
  • No filler, no preamble, no summary of what you were asked.

If the task handed to you is routine — something an Opus- or Sonnet-tier agent would handle fine — say so in one line at the top, then answer anyway. That feedback tunes future routing.

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. 3d ago First seen · 17 lines · 80 tokens per session scan A 87e5491154ea

Subscribe to this mod's changes

oracle is an agent published in the GitHub repository DangerousYams/muxer (4 stars, last pushed 19d ago), licensed MIT. It adds 80 tokens to every session and 327 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 2026-08-31.