agent-concierge-probe

agent-concierge-probe is a skill for Claude Code from The-Utopia-Studio/skills. It costs 204 tokens per session (1,686 once invoked), scanned A, original, Apache-2.0.

A measured trial in which an AI agent performs a real, already-mapped task from start to finish across several runs. It records which steps need human help and the cost of each result.

In plain words
What is it for?
It helps measure input and output tokens, tool calls, human-fix time and cost per outcome, then map the steps the agent can or cannot complete alone.
Why use it?
It replaces a single successful demonstration or guess with evidence about automation limits, human fixes and the resources used.

Skill for Claude Code

Written for Claude Code: shipped in a Claude Code plugin. Also seen: positional $N argument.

Part of the utopia-product plugin — 13 skills shipped together

Good fit It helps measure input and output tokens, tool calls, human-fix time and cost per outcome, then map the steps the agent can or cannot complete alone.

Compare 6 skills from other repositories ↓
Install with agentmods
npx agentmods add skills/the-utopia-studio/skills/agent-concierge-probe
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.

Any agent
npx skills add The-Utopia-Studio/skills --skill agent-concierge-probe
Clone the repo
git clone --depth 1 https://github.com/The-Utopia-Studio/skills

Made for: Claude Code.

Or install utopia-product, the plugin that ships this one along with the rest of its 13 skills.

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 agent-concierge-probe

README.md
[![agentmods](https://agentmods.dev/badge/skills/the-utopia-studio/skills/agent-concierge-probe/github.svg)](https://agentmods.dev/skills/the-utopia-studio/skills/agent-concierge-probe)
Your own site
<a href="https://agentmods.dev/skills/the-utopia-studio/skills/agent-concierge-probe"><img src="https://agentmods.dev/badge/skills/the-utopia-studio/skills/agent-concierge-probe/github.svg" alt="Measured on agentmods" height="20"></a>

Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.

agentmods 80×15 button for agent-concierge-probe

Your own site · 80×15
<a href="https://agentmods.dev/skills/the-utopia-studio/skills/agent-concierge-probe"><img src="https://agentmods.dev/badge/skills/the-utopia-studio/skills/agent-concierge-probe.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 204 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 1,686 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.
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 $0.00204 $0.01686
Opus 5 $0.00102 $0.00843
Sonnet 5 $0.00041 $0.00337
Haiku 4.5 $0.00020 $0.00169

Measured 4d ago against content hash 0bef90511191, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-09, from the pricing page.

Security

Grade A, and why

agent-concierge-probe 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 4d 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.

plugins/utopia-product/skills/agent-concierge-probe/SKILL.md · 89 lines

How it starts

The opening of the file, as written. The whole thing — 89 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Agent Concierge Probe

What it does

You hand a real, already-mapped task to an agent, let it run the task end to end across several outcomes, and instrument every run. Output is two things: the cost-per-outcome priced to the cent, and a frontier map showing which steps the agent does unaided and which a human has to take over. It compresses "is this automatable, and what does one outcome cost" into measured numbers instead of a hunch.

The Icarus reframe

A generic "can an agent do it?" demo shows one happy-path run and a vibe. This runbook refuses a cost claim without its parts: the read-out must show input tokens, output tokens, tool calls, and human-fix minutes separately, each priced, before it states a cost-per-outcome — a bare "$2/task" is not a result. And it treats the red steps (where a human had to intervene) as the actual finding: those steps are the automation frontier, the line the product cannot yet cross. The probe is honest about cost-per-outcome and the frontier. It lies about self-serve UX and about trust/adoption, and it says so in every read-out.

When to use / When NOT

Use it when the workflow is already mapped (ideally by a concierge run, so you know the exceptions) and you need to know what an agent can do unaided and what one outcome costs.

Do not use it when:

  • You have not chosen a probe — that is probe-matrix.
  • You want to test willingness to pay, or you (the founder) will do the task by hand — that is concierge-probe.
  • You want to know whether people would use or trust the thing — this probe lies about both. Adoption and trust need a concierge or field probe.
  • You want to fake an interface to test would-they-use-it — that is wizard-of-oz-probe.
  • You want to test whether a workflow reads / is legible on paper before anyone builds it — that is paper-sketch-probe.

Method

Follow the runbook. Fill template.md as you go.

Step 1 — Take a mapped workflow. Break the task into discrete steps. If a concierge run already exposed the exceptions, carry them in — the agent will hit them.

Read the full file on GitHub · 89 lines

Files

What ships with it

12 files 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. 4d ago First seen · 89 lines · 0 tokens per session scan A 0bef90511191

Subscribe to this mod's changes

agent-concierge-probe is a skill published in the GitHub repository The-Utopia-Studio/skills (4 stars, last pushed 6d ago), licensed Apache-2.0. It adds 204 tokens to every session and 1,686 once invoked, about $0.0010 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 2026-09-05.

Related

Other skills, from other repositories

find-skills

Discovers, searches, and installs skills from multiple AI agent skill marketplaces (400K+ skills) using the SkillKit CLI. Supports browsing official partner collections (Anthropic, Vercel, Supabase, Stripe, and more) and community repositories, searching by domain or technology, and installing specific skills from…

rohitg00/skillkit · 162 tokens

github

When the user wants to use GitHub for SEO, parasite SEO, GEO, open source marketing, README optimization, or curated Awesome lists. Also use when the user mentions "GitHub," "GitHub SEO," "GitHub parasite SEO," "GitHub GEO," "awesome list," "GitHub README," "profile README," "pinned repositories," "Trending,"…

kostja94/marketing-skills · 165 tokens

brand-visual-generator

When the user wants to define, audit, or apply visual identity (typography, colors, spacing, design tokens, frontend aesthetics). Also use when the user mentions "brand style guide," "visual identity," "design system," "typography," "color palette," "brand guidelines," "AI brand aesthetics," "brand colors," "font…

kostja94/marketing-skills · 124 tokens

image-optimization

When the user wants to optimize images for search engines and performance. Also use when the user mentions "image SEO," "alt text," "image captions," "figcaption," "image optimization," "WebP," "lazy loading," "LCP," "image sitemap," "responsive images," "srcset," "image format," or "hero image optimization." For CWV…

kostja94/marketing-skills · 87 tokens

xml-sitemap

When the user wants to create, audit, or optimize sitemap.xml. Also use when the user mentions "sitemap," "sitemap.xml," "sitemap index," "lastmod," "changefreq," "priority," "URL discovery," "URL discovery for search engines," "single source of truth," "URL config," "unify sitemap IndexNow," or "reduce duplicate…

kostja94/marketing-skills · 91 tokens

robots-txt

When the user wants to configure, audit, or optimize robots.txt. Also use when the user mentions "robots.txt," "crawler rules," "block crawlers," "AI crawlers," "GPTBot," "allow/disallow," "disallow path," "crawl directives," "user-agent," "block Googlebot," "fix robots.txt," "robots.txt blocking," or "search engine…

kostja94/marketing-skills · 91 tokens