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.
npx agentmods add skills/zgbrenner/agentcounsel/worker-classificationnpx skills add zgbrenner/agentcounsel --skill worker-classificationgit clone --depth 1 https://github.com/zgbrenner/agentcounselWrote 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.
[](https://agentmods.dev/skills/zgbrenner/agentcounsel/worker-classification)<a href="https://agentmods.dev/skills/zgbrenner/agentcounsel/worker-classification"><img src="https://agentmods.dev/badge/skills/zgbrenner/agentcounsel/worker-classification.svg" alt="Measured on agentmods" height="20"></a>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.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5 | $0.00043 | $0.03308 |
| Opus 5 | $0.00022 | $0.01654 |
| Sonnet 5 | $0.00009 | $0.00662 |
| Haiku 4.5 | $0.00004 | $0.00331 |
Grade A, and why
Worker Classification 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.
How it starts
The opening of the file, as written. The whole thing — 162 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Worker Classification
Purpose
Produce a structured, attorney-ready classification analysis memo for a proposed worker engagement — one that has not yet started. The skill gathers and organizes the factual inputs that bear on worker classification, identifies the test or tests that apply in the governing jurisdiction for the relevant legal purpose, applies those tests to the facts in a structured factor table, and flags gaps between the intended arrangement and what the facts support. It produces draft legal work product for attorney review — not legal advice, not a legal conclusion on the lawfulness of any classification, and not a substitute for employment counsel.
Use When
- A company or HR professional is planning a new engagement with an individual worker and needs to think through whether the arrangement is supportable as an independent contractor relationship.
- An attorney needs a structured intake of engagement facts to begin a classification risk assessment before work begins.
- A user asks "can we bring this person on as a contractor?" or "what factors should we look at before we classify this engagement?"
- A proposed arrangement involves a staffing agency, vendor SOW, or direct contractor agreement, and classification risk has not yet been assessed.
- The user needs to compare the intended contract structure against the operational facts before the engagement is executed.
Required Inputs
- Description of the work: what the worker will do, day to day; whether the work is core to the company's business or peripheral; whether the engagement is project-based or indefinite in duration; and the level of specialization or skill required.
- Control facts: whether the company will set the worker's hours or schedule; whether the work will be performed on company premises; who will direct the method and sequence of work; whether the company will have supervisory authority over the worker.
- Economic facts: how the worker will be paid (flat fee, hourly, milestone, retainer); who will provide tools, equipment, and materials; whether the worker has or will have other clients during the engagement; and whether the worker bears any risk of profit or loss on the engagement.
- Arrangement structure: whether this is a direct independent contractor agreement, a staffing-agency placement, or a vendor SOW; proposed duration; and whether the worker will be physically co-located with employees.
- Classification purpose(s): the legal purpose(s) for which classification matters (e.g., income tax withholding, wage-and-hour obligations, unemployment insurance, benefits eligibility, workers' compensation coverage). Different purposes may trigger different tests under the governing jurisdiction's law.
- Jurisdiction: the state, province, or country whose law will govern the engagement. If unknown, flag as
[CONFIRM: governing jurisdiction].
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.
- 4d ago First seen · 162 lines · 43 tokens per session scan A 794d71a797a0
Worker Classification is a skill published in the GitHub repository zgbrenner/agentcounsel (18 stars, last pushed 1mo ago), licensed MIT. It adds 43 tokens to every session and 3,308 once invoked, about $0.0002 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.
Other skills, from other repositories
opencaselaw-maintenance
Use for OpenCaseLaw autonomous maintenance loops, production health checks, scraper/completeness triage, safe deploy decisions, and Codex automation work in this repo. Do not use for ordinary feature work unless the task is about operational safety, corpus completeness, or maintenance automation.
format-monitoring-report
Оформи утверждённое содержание правового мониторинга в DOCX или PDF по семисекционному шаблону с синими плашками, Times New Roman, сносками, нумерацией страниц и итоговой сравнительной таблицей. Используй только для структуры и визуального оформления готового текста; не используй для исследования, методологии…
verifiable-legal-document
Собери проверяемый юридический документ: claim IDs, кликабельные внутренние ссылки на точные предложения/абзацы источников в приложениях, hashes и автоматический validation report. Используй для итогов высокой важности; не используй для неподтвержденного черновика.
matter-intake
Инициализируй новое юридическое дело или исследование: классификация данных, вопрос, дата актуальности, границы, продукт, план и evidence gate. Используй в начале нового поручения; не используй для уже полностью scoped задачи.
ukrainian-legal-research
Исследуй действующее право Украины на конкретную дату: иерархия, редакции, вступление в силу, переходные положения, компетенция и официальные источники. Используй для украинского нормативного вопроса; не используй для написания финального документа без evidence ledger.
legal-qa-audit
Проведи независимую юридическую и доказательственную проверку черновика: применимость норм, точность цитат, adverse authority, числа, overclaiming, ссылки, приватность и готовность. Используй перед выдачей; не используй как самоодобрение автора.