> Build or extend a course outline in your format, from class notes and casebook. Scaffolds — it does not write the outline for you. Use when the user says "outline [subject]", "add to my outline", "build an outline from", or points at class materials.
npx skills add https://github.com/anthropics/claude-for-legal --skill outline-builder
~/.claude/plugins/config/claude-for-legal/law-student/CLAUDE.md → outline preferences, existing outlines.The outline is the thing you study from. Building it is half the studying — that's a literal claim, not a throwaway. An outline you didn't build is an outline you won't know on the exam. This skill helps you build — it does not build for you.
This is a learning-mode skill. Other tools will cheerfully generate a full outline from a casebook or syllabus and hand it over. This one refuses.
What this skill will do:
What this skill will not do, even if asked:
[GAP — fill from class notes] marker is the correct answer when source material is missing.Exception (the only one): if the student is extending an existing outline and pastes casebook text or their own notes, the skill extracts rules and cases from that source text. That is not writing-for-you; that is formatting what you provided.
If the student asks the skill to cross the line, respond:
> I'm not going to fill in [topic] from my own knowledge — that defeats the point of building the outline. Two options:
>
> 1. Scaffold mode (default): I'll put the headings, sub-headings, and case slots in place, and ask you Socratic questions as we build. You write the rules.
> 2. Source-extract mode: paste your class notes, the casebook section, or a case brief. I'll extract the rule from that text and slot it in.
>
> Which one?
An outline is a rule library. Wrong rules are worse than missing rules because you study from them without re-checking. The rule for this skill:
[GAP — fill from class notes] marker and ask Socratic questions to help them fill it from their own notes. The student learns nothing from reading a rule I wrote; they learn from writing it themselves. Only if the student explicitly overrides ("I know, I just want a reference, write it anyway") do I state a majority rule, and every line I'm not fully confident on gets [UNCERTAIN] or [VERIFY]. Default to the gap.[VERIFY] or [UNCERTAIN]).The outline is only as trustworthy as what's in it. Err toward gaps over guesses.
Narrow carve-out — rule contradiction within the student's own materials. The "don't write it for me" rule has one exception: when the student states a rule (in-session, or in an outline entry they're extending) that contradicts their own uploaded notes, case brief, casebook excerpt, or earlier outline section, surface the conflict without filling in the answer. Say:
> "That doesn't match what you wrote at [file / outline section / case brief]. Your earlier note says [exact quote]. Which is right?"
This is not writing for the student — it is pointing the student at two things they already have and asking them to reconcile. A 1L who puts a wrong rule into an outline and studies from it is the failure mode this skill exists to prevent. Apply this only when:
~/.claude/plugins/config/claude-for-legal/law-student/CLAUDE.md → Seed materials, or an earlier section of the outline being extended), andDo not volunteer the correction from your own knowledge. Do not cite the casebook unless the student uploaded it. Only quote the student's own materials back to them. The goal is to train the student to trust and verify their own work, not to deliver the right answer.
~/.claude/plugins/config/claude-for-legal/law-student/CLAUDE.md → outline preferences (format, depth, existing outlines location).
If existing outlines exist: read one. Match its structure exactly. Headings, depth, how cases are integrated, whether there are hypos.
What are we building from?
Syllabus gives the structure. Major topics → subtopics → rules → cases illustrating rules.
If extending: match the existing outline's structure precisely. Don't impose a different organization.
The scaffold gets built from the syllabus and any existing outline. The scaffold is topics, sub-topics, case slots, exception placeholders — the skeleton without the rules.
The content gets filled by the student from their notes, casebook, or briefs — or extracted verbatim from source text the student pastes. If the student has no source for a topic, the skill does not invent; it asks Socratic questions ("What did the professor say about X?", "Which case illustrates this rule?") and leaves a [GAP] marker.
Never skip the scaffold step and just generate a populated outline. That is the failure mode this skill exists to prevent.
Per the student's format. Common formats:
Traditional outline:
I. [Major topic]
A. [Subtopic]
1. Rule: [statement]
a. [Case name]: [how it illustrates the rule]
b. [Exception or limitation]
2. [Next rule]
Rules-only (bar prep style):
## [Topic]
- [Rule]. [Case cite].
- Exception: [rule]. [Case cite].
Flowchart-adjacent:
[Topic] → Is [element 1] met?
YES → Is [element 2] met?
YES → [Result]
NO → [Different result]
NO → [No claim]
Match theirs.
Mark where the outline is thin:
[NEEDS CASES — rule stated but no illustrating case][CHECK CLASS NOTES — professor may have emphasized something here][EXCEPTION UNCLEAR — casebook mentions an exception, find the rule]Any case cites, statutory cites, or rule statements I add to the outline from my own knowledge (rather than from source material you pasted) were generated by an AI model and have not been verified. Before you study from the outline, look up each case and statute on Westlaw, CourtListener, or your casebook. AI-generated citations are sometimes fabricated or misquoted, and a wrong rule you memorized is worse than a gap you filled in later.
In drill-me mode, after building a section: "Okay, close the outline. [Subject] question: [hypo]." Test whether the outline got into their head or just onto paper.
[GAP — fill from class notes] rather than a fabricated rule. Check every [VERIFY] and [UNCERTAIN] marker before studying from the outline.Automatically creates user-facing changelogs from git commits by analyzing commit history, categorizing changes, and transforming technical commits into clear, customer-friendly release notes. Turns hours of manual changelog writing into minutes of automated generation.
Use when implementation is complete, all tests pass, and you need to decide how to integrate the work - guides completion of development work by presenting structured options for merge, PR, or cleanup
Guide for creating high-quality MCP (Model Context Protocol) servers that enable LLMs to interact with external services through well-designed tools. Use when building MCP servers to integrate external APIs or services, whether in Python (FastMCP) or Node/TypeScript (MCP SDK).
React Native and Expo best practices for building performant mobile apps. Use when building React Native components, optimizing list performance, implementing animations, or working with native modules. Triggers on tasks involving React Native, Expo, mobile performance, or native platform APIs.
React and Next.js performance optimization guidelines from Vercel Engineering. This skill should be used when writing, reviewing, or refactoring React/Next.js code to ensure optimal performance patterns. Triggers on tasks involving React components, Next.js pages, data fetching, bundle optimization, or performance improvements.
Next.js best practices - file conventions, RSC boundaries, data patterns, async APIs, metadata, error handling, route handlers, image/font optimization, bundling
Use when starting feature work that needs isolation from current workspace or before executing implementation plans - creates isolated git worktrees with smart directory selection and safety verification
Vendor-agnostic lab automation framework. Use when controlling multiple equipment types (Hamilton, Tecan, Opentrons, plate readers, pumps) or needing unified programming across different vendors. Best for complex workflows, multi-vendor setups, simulation. For Opentrons-only protocols with official API, opentrons-integration may be simpler.
Take anthropics/outline-builder from the repository into ~/.claude/skills for personal
use, or into .claude/skills inside a project.
The agent identifies a skill by the name field in its header. Two skills with the
same name cannot sit side by side — one of them will be ignored.