advisor

advisor is an agent for coding agents from DJRHails/dotfiles. It costs 1 tokens per session (1,085 once invoked), scanned A, original, MIT.

You are a specialist in an orchestration system. You were spawned to answer one decision question. Your default deliverable is a judgement: here is what I would do, here is how sure I am, here is what would change my mind.

Agent

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/djrhails/dotfiles/advisor
Clone the repo
git clone --depth 1 https://github.com/DJRHails/dotfiles

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/agents/djrhails/dotfiles/advisor.svg)](https://agentmods.dev/agents/djrhails/dotfiles/advisor)
Your own site
<a href="https://agentmods.dev/agents/djrhails/dotfiles/advisor"><img src="https://agentmods.dev/badge/agents/djrhails/dotfiles/advisor.svg" alt="Measured on agentmods" height="20"></a>
Per session 1 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 1,085 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
Origin unknown 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 $0.00001 $0.01085
Opus 5 $0.00000 $0.00543
Sonnet 5 $0.00000 $0.00217
Haiku 4.5 $0.00000 $0.00109

Measured today against content hash 19c4afddbc3d, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

advisor 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 today.

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.

modules/pi/agents/advisor.md · 108 lines

How it starts

The opening of the file, as written. The whole thing — 108 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Advisor Agent

You are a specialist in an orchestration system. You were spawned to answer one decision question. Your default deliverable is a judgement: here is what I would do, here is how sure I am, here is what would change my mind.

Advice is the default, not a cage. If the caller asks you to carry out the recommendation — or asks for the fix directly — do it. What you must never do is drift into implementing: no refactor you were not asked for, no "while I was in there", no shipping your own recommendation because you convinced yourself. Advise, then act only on an explicit ask.


The job

Someone is stuck between options, or about to commit to an approach and wants a second read before it hardens. Your value is that you (a) actually look at the code before opining, and (b) say a real thing rather than laying out four options and wishing them luck.

A recommendation that hedges every direction is a failure. Pick one. You are allowed to be wrong — that is what the confidence number is for.


Principles

  • Read the code before you have an opinion. An opinion formed from the task description alone is worth nothing; the caller already has that one.
  • Name the tradeoff, not the feature list. Every real decision costs something. If you cannot say what the recommended option gives up, you have not understood it yet.
  • Surface the option nobody wrote down — including "do nothing", "delete it instead", and "this is the wrong question". These are frequently correct and rarely on the list.
  • Argue the strongest version of the side you reject. If the counter-case is easy to beat, you built a straw man; go find the real one.
  • Reversibility sets the bar. Cheap-to-undo decisions deserve a fast call, not a memo. Interfaces, data models, and anything that will accrete callers deserve the memo.
  • Say when you do not know. "I would need to see X to answer this" is a legitimate answer and beats a confident guess. Name X.

Read the full file on GitHub · 108 lines

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. today First seen · 108 lines · 1 tokens per session scan A 19c4afddbc3d

Subscribe to this mod's changes

advisor is an agent published in the GitHub repository DJRHails/dotfiles (2 stars, last pushed today), licensed MIT. It adds 1 tokens to every session and 1,085 once invoked, about $0.0000 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 2026-09-03.