Field Onboarding skill
A skill that guides a researcher into a new field by rungs: motivation, vocabulary, methods, frontier, no unverified references
- 80 ★ on the repository
- Last commit
Minimal access
Why this level
- The skill runs a dialogue and checks references when search is available, and where the agent can run commands it runs a local knowledge state script
Install
- Works in
In your terminal, with SkillFoxx CLI
npx skillfoxx add skills/field-onboardingDetects the agents on your machine, shows the access level and pins the version.
Other ways to install
Access and checks
How we reviewMinimal accessThe tool runs nothing and makes no network calls.
- The skill runs a dialogue and checks references when search is available, and where the agent can run commands it runs a local knowledge state script
This is third-party code. Review the repository files before installing.
What it does
Field Onboarding is a skill that guides a researcher into an unfamiliar field and decodes a paper, abstract, figure caption or referee comment. The agent first finds the field structure, takes three prerequisites from it, asks which you already have and sets your goal. Teaching then goes in rungs: why the field exists, a vocabulary of five to ten terms with symbols and false friends, then one turn per part of the field, and finally open questions and a reading path. The goal routes the whole path: reading a paper stresses notation and formalism, judging a method stresses phenomena and worked numbers, hands-on work stresses procedure. The output is a takeaway: a map with the state of each part, a glossary, a reading path and what remains uncovered. For supplied text a Decode mode separates source claim, background, inference and critique.
Who it is for. Graduate students and researchers moving into an adjacent field who want to understand it without a specialist-level recap.
Good fit when
- You know nonlinear optics and want to enter topological photonics step by step
- A paper has unclear notation and an abstract that must be taken apart
- You need a reading list marked with what was verified and what was named from memory
- You read a referee report and want to see what is critique and what is background
- Before a seminar you need a field map and a list of questions still open
Not a fit when
- You need a short factual answer, such as what the abbreviation PL means
- You are a specialist in exactly this area asking a narrow technical question
- You asked for a one line answer: a full rung path only gets in the way
- You need a ready literature review for publication: the skill teaches rather than writes a review
Example request
I understand nonlinear optics but not topological photonics. Guide me into the field step by stepLimitations
The skill remembers nothing between sessions: you save the closing takeaway with the map and paste it into a new session. Paper references are split into those verified in this session with a DOI, arXiv ID or journal and those named from memory unverified, and bare unlabelled references are not allowed. Where verification is impossible it gives an executable search pointer instead of a reference. Capable agents use the bundled `knowledge_state.py` script themselves and the user never runs it. Agents without command execution follow the same prompt workflow. The mode does not trigger for narrow factual questions.
How to disable. Delete the `field-onboarding` folder from the agent skills directory. Saved takeaways live with you separately and remain.
FAQ
How do I install the Field Onboarding skill?
Run `npx skills add ljx-chase/research-field-onboarding -g` for Claude Code. For other agents ask for installation per the `AGENTS.md` section, and for Claude web and ChatGPT a zip sits in releases.
How does the conversation start?
With a question about prerequisites and goal. The agent finds the field structure, takes three prerequisites from it and asks which you already have. It does not open with a paragraph of definitions.
How does the skill handle paper references?
Those verified in the session get a DOI, arXiv ID or journal and others are explicitly labelled as from memory, unverified. Bare unlabelled references are not allowed.
What is Decode mode?
It switches on for supplied text: a paper, abstract, figure caption or referee comment. Source claim, background, inference and critique are separated so you can see which is which.
How do I continue in a new session?
Save the closing takeaway with the map and paste it into a new session. The skill remembers nothing between sessions on its own.
Discussion
No questions or reviews yet.
Related
SkillsField Lab offers instruments for examining a question: a term scan, an event map, a 2×2 matrix and a field log of observations.
A skill that turns a vague interest, a literature gap or a proposal draft into a testable research question with a first-step plan
Readwise skills help agents triage a reading inbox, discuss books and connect highlights across saved sources
A job-hunt skill bundle: JD decoding, template resumes, behavioral interview prep, offer comparison and negotiation
A skill set for resumes and job search: tailoring to a vacancy, ATS pass-through checks, cover letters and interview preparation
get-job.skill (实习.skill)
get-job.skill
A skill that prepares a job search: it researches the role, rewrites your resume and drills you on every bullet
Comparison with similar tools
Cards that share the main search query, and how they differ.
| Tool | Official | Repository stars | Main difference |
|---|---|---|---|
| Field Onboarding (this card) | no | 80 | |
| tutor-skills | no | 1,402 | general tutoring skills; here rungs for entering a research field with a map and reference checking |
| DeepTutor CLI Skill | no | 58,697 | a tutor CLI; here a skill with a goal-routed path and a paper decode mode |
README in short
The README explains how the skill differs from an ordinary specialist answer to a beginner and shows a quick install and a trial prompt. It then covers the learning rungs, the map and path through the field, reference checking rules and the mode for decoding supplied text.
README badge
Original SKILL.md
--- name: field-onboarding description: Guide a researcher step by step into an unfamiliar research field, or decode a paper, abstract, figure caption, or referee comment they cannot parse. Builds understanding in rungs (motivation, vocabulary, core framework, methods, frontier), anchored to what the user already knows, with a checkpoint before each advance. Use when the user says they are new to a field, asks what a research area or method is, says an explanation was too technical, asks to be walked through something step by step, asks for a reading path, or supplies dense research text. Trigger even when the user only names an unfamiliar field or pastes an abstract without asking to be taught. Do not trigger for a one-line summary, a short overview request, or a direct factual answer; do not use it for translation, editing, search, debugging, or specialist questions inside the user's own field. Also trigger in other languages, including Chinese such as 入门, 一步一步讲, 看不懂, 这篇论文讲什么, 帮我理解这个领域. --- # Field Onboarding Get someone productively oriented in an unfamiliar research field, fast, without losing them. The failure this skill exists to prevent: answering a beginner's question at the level of a specialist, so the answer is technically correct and completely useless. ## When not to use this skill This is a teaching mode, not a default. Running the full ladder on someone who wanted one sentence is its own failure, and a more irritating one than pitching too high. Do not run Step 0 or the ladder when: - **The question is narrow and factual.** "What does PL stand for?" "What wavelength do people usually pump at?" Answer it. Do not calibrate. - **The user is already a specialist in this exact area** and is asking a specific technical question inside it. - **The user asked for it short.** "quickly", "one line", "just tell me", "TL;DR", "简单说", "赶时间", "Give me a two-sentence overview of X", or "summarize the field in a paragraph". These are not onboarding requests unless the user explicitly says they want to be taught step by step. - **The task is not understanding.** Translation, proofreading, formatting, debugging, writing, or a literature search with a known target. - **The user is mid-task and blocked.** Someone whose fit is failing at 2am needs the fix, not motivation and vocabulary. - **The user already declined the ladder this session.** Offer once. Never twice. **The one-turn test.** Before starting Step 0, ask whether the question can be answered well in a single turn. If it can, answer it, then offer the ladder in one line: "that is the short answer; if you want to actually work in this area, I can walk you up from the motivation." Offer once, drop it if unclaimed. When in doubt, answer first and offer second. A good answer followed by an offer costs the user nothing. An intake questionnaire in front of a one-line question costs them a turn and their patience. ## Core rules