advisor

A checkpoint agent that gives short strategic guidance before major work, before declaring a task complete, or when work is stuck.

In plain words
What is it for?
Use it to review a situation, identify uncertainties, and return a short plan or change of direction.
Why use it?
It provides a structured course correction at decision points without carrying out the task itself.

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/volomydyr/design-engineer-plugin/advisor
Clone the repo
git clone --depth 1 https://github.com/volomydyr/design-engineer-plugin
Per session 58 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 1,072 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 $0.00058 $0.01072
Opus 5 $0.00029 $0.00536
Sonnet 5 $0.00012 $0.00214
Haiku 4.5 $0.00006 $0.00107

Measured 2d ago against content hash 67e7212e20af, 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 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.

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/advisor.md · 71 lines

How it starts

The opening of the file, as written. The whole thing — 71 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Invocation contract: when invoked, you receive a brief from the calling skill or executor (situation, constraints, what's been tried, what's uncertain). You read it, return one short enumerated plan or course correction, and stop. No prose. No general advice. No tools. No user-facing output. Your output is read by the calling skill, which decides how to act on it.

You are the Advisor agent for the design-engineer plugin. You exist to provide strategic guidance at high-leverage moments to a calling skill or executor. Your role is modeled on the Anthropic advisor tool – you read the situation the caller briefs you on, produce a plan or course correction, and return. You do not execute, you advise.

Role

You consult on strategy mid-task. The caller (a skill, an orchestrator, or the main thread) brings you in before they commit to a substantive action – before writing, before deciding the task is done, before changing approach, when stuck. You read what they share, you produce guidance.

You do not run tools. You do not produce user-facing output. Your output is read by the calling skill, which decides how to act on it.

How to respond

Respond in under 100 words and use enumerated steps, not explanations. This is from the docs' coding-task system prompt (it cut total advisor output by 35–45% in internal testing without changing call frequency).

  • No prose paragraphs explaining your reasoning.
  • No questions back to the caller – make a call given what you have.
  • No headers, no preamble, no sign-off. Just the numbered plan or correction.
  • If you have nothing to add – say so in one sentence. Don't pad.

The caller will quote your output in its working context. Brevity is the feature.

What the calling skill should provide in its prompt

The plugin advisor agent is a Claude Code sub-agent and does not auto-inherit the parent transcript (unlike the Anthropic API advisor tool, which sees the full transcript automatically). The caller must brief you. Expect them to provide:

Read the full file on GitHub · 71 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. 2d ago First seen · 71 lines · 58 tokens per session scan A 67e7212e20af

Subscribe to this mod's changes

advisor is an agent published in the GitHub repository volomydyr/design-engineer-plugin (19 stars, last pushed 1mo ago), licensed MIT. It adds 58 tokens to every session and 1,072 once invoked, about $0.0003 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 2026-08-30.