advisor

advisor is a skill for Claude Code, Codex from KevinZai/commander. It costs 71 tokens per session (2,603 once invoked), scanned A, original, MIT.

A model-pairing tool where a faster, cheaper model handles a task and consults a more capable model when it gets stuck or needs a check. Both models work within one API request.

In plain words
What is it for?
Use it for difficult code generation, refactoring, or analysis when you want most work handled by a cheaper model with occasional senior-level guidance.
Why use it?
It helps with complex, multi-step work where constant use of the more capable model would cost more. The executor can get guidance without requiring separate orchestration code.

Skill for Claude CodeCodex

Which agent this was written for is unclear — built for openclaw. Also seen: positional $N argument; built for openclaw.

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 skills/kevinzai/commander/advisor
Any agent
npx skills add KevinZai/commander --skill advisor
Clone the repo
git clone --depth 1 https://github.com/KevinZai/commander

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 advisor

README.md
[![agentmods](https://agentmods.dev/badge/skills/kevinzai/commander/advisor.svg)](https://agentmods.dev/skills/kevinzai/commander/advisor)
Your own site
<a href="https://agentmods.dev/skills/kevinzai/commander/advisor"><img src="https://agentmods.dev/badge/skills/kevinzai/commander/advisor.svg" alt="Measured on agentmods" height="20"></a>
Per session 71 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 2,603 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 1 finding. 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 5.1 $0.00071 $0.02603
Opus 5 $0.00036 $0.01301
Sonnet 5 $0.00014 $0.00521
Haiku 4.5 $0.00007 $0.00260

Measured 2d ago against content hash a5b0da3d6beb, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-06, from the pricing page.

Security

Grade A, and why

advisor 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 2d 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.

curl https://api.anthropic.com/v1/messages \
skills/advisor/SKILL.md · 327 lines

How it starts

The opening of the file, as written. The whole thing — 327 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Anthropic Advisor Tool

The Advisor Tool lets a faster, cheaper model (the executor — Haiku or Sonnet) run a task while silently consulting a smarter model (Opus) for guidance whenever it gets stuck or needs a sanity check. Everything happens inside a single /v1/messages request: the executor decides when to call the advisor like any other tool, Opus sees the full conversation and returns advice, and the executor continues. No extra round trips, no separate API calls, no special orchestration code on your end.

Think of it as giving Sonnet a senior engineer it can quietly ask for help — without paying Opus rates for the whole job.


When to Use It

Good fit:

  • Long, multi-step tasks where Sonnet handles most steps but needs occasional guidance on hard parts
  • Complex code generation, refactoring, or analysis where quality matters but budget is a concern
  • Any task where "Sonnet with supervision" beats "Opus for everything" on cost

Poor fit:

  • Single-turn Q&A — the overhead isn't worth it for one-shot requests
  • Simple pass-through tasks where Sonnet already performs well
  • Cases where you need the advisor's full reasoning chain (thinking blocks are stripped; only the advice text reaches the executor)

Model Compatibility

The rule (per Anthropic's advisor tool docs): the advisor must be Claude Sonnet 4.6 or more capable, AND at least as capable as the executor.

Both conditions must hold:

  1. The advisor is Claude Sonnet 4.6 or more capable.
  2. The advisor is at least as capable as the executor — an advisor can't be weaker than the model it's advising.

🔗 Check the exact per-executor matrix in Anthropic's docs before pinning a pair. That table is deliberately not mirrored here — a copied matrix goes stale silently every time Anthropic ships a model, and a wrong row sends you a 400 at runtime.

Read the full file on GitHub · 327 lines

Files

What ships with it

1 file beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.

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. 2d ago First seen · 327 lines · 71 tokens per session scan A a5b0da3d6beb

Subscribe to this mod's changes

advisor is a skill published in the GitHub repository KevinZai/commander (6 stars, last pushed yesterday), licensed MIT. It adds 71 tokens to every session and 2,603 once invoked, about $0.0004 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 2026-09-04.