Skip to main content
skill CX Codex official · Anthropic

session — Claude Code skill: what it does, how to install

Runs a fixed-length study session on one subject (MBE questions, essays or flashcards) and writes the score back into the study plan.

Install into Claude Code · Skills CLI (Vercel) All methods and platforms ↓
npx skills add anthropics/claude-for-legal --skill session -a claude-code
Open on GitHub

What session does

A study session on one subject, for a fixed number of questions. The skill parses the arguments first: subject and count. If either is missing it asks. Then it reads ~/.claude/plugins/config/claude-for-legal/law-student/CLAUDE.md for jurisdiction, exam format and weak subjects, plus study-plan.yaml when there is one, so questions lean toward the subtopics that went badly before.

A flag picks the method. --mbe runs bar-style multiple choice and is the default for bar prep subjects. --essay gives written prompts graded against an essay rubric. --flashcards drills cards. Questions come one at a time with an explanation after each, and where jurisdictions diverge the answer is tagged [UBE/majority] or [state-specific] with the rule body spelled out.

Results land in session_history inside study-plan.yaml, or in session-history.yaml in the same config folder when no plan exists. That history is what lets the plan adapt, so a missing CLAUDE.md means no read on the student's weak areas. Two other skills come along: --mbe and --essay call bar-prep-questions, --flashcards calls flashcards.

What it helps with

  • drill a block of MBE questions on a subject that keeps slipping
  • grade an essay prompt against the rubric and see what was missing
  • run flashcards as a drill instead of a flip-through
  • keep a record of each session's score
  • get the next topic from a plan that accounts for earlier misses

Example prompts for the agent

  • /session Evidence <n> --mbe. Explain after each question why the other options fail.
  • Run /session on Contracts for <n> essays and grade them against the rubric.
  • /session Torts <n> --flashcards, then update the study plan.

How to install session

1Skills CLI (Vercel)

npx skills add anthropics/claude-for-legal --skill session -a claude-code

2Manually: put the files in place

~/.claude/skills/session/SKILL.md

Copy the whole skill folder; for one project use .claude/skills/session/

Step by step, with commands for each platform: How to install skills in Claude Code: five ways

Questions about session

How do I install session in Claude Code?

Run npx skills add anthropics/claude-for-legal --skill session -a claude-code. The skill file ends up at ~/.claude/skills/session/SKILL.md.

What does the session skill do?

It runs a set number of questions on a single subject, as MBE items, essays or flashcards, explains each answer and records the score so the study plan can shift toward weak subtopics.

What do I need for it to work?

The CLAUDE.md config in the law-student plugin folder, which supplies jurisdiction, exam format and weak subjects. The MBE and essay modes also call the bar-prep-questions skill, and flashcard mode calls flashcards.

Source text

SKILL.md · Apache-2.0
---
name: session
description: >
  Run a focused N-question study session on a subject — MBE, essay, or
  flashcards. Tracks performance and updates the study plan. Use when the
  user says "run me 10 questions on [subject]", "do a session on [subject]",
  "let's do 5 cards on [subject]", or wants to drill a fixed number of
  questions and have the plan adapt.
argument-hint: "<subject> <n> [--mbe | --essay | --flashcards]"
---

# /session

1. Parse `$ARGUMENTS` — subject and N. If missing, ask:
   > What subject, and how many questions? (e.g., `Evidence 10` or `Contracts 5 --essay`.)
2. Load `~/.claude/plugins/config/claude-for-legal/law-student/CLAUDE.md` → jurisdiction, exam format, weak subjects.
3. Load `~/.claude/plugins/config/claude-for-legal/law-student/study-plan.yaml` if it exists. Read `session_history` for this subject to weight subtopics toward where the student has been weak.
4. Route by method flag:
   - `--mbe` (default for bar prep subjects): load `bar-prep-questions` skill, run N MBE-style questions. Apply jurisdiction handling (see that skill's `## Jurisdiction handling`). Label each `[UBE/majority]` or `[state-specific]`.
   - `--essay`: load `bar-prep-questions`, run N essay prompts. Grade per essay-mode rubric.
   - `--flashcards`: load `flashcards` skill, run N cards in `--drill` mode.
5. Run N questions one at a time. After each, explain right/wrong and flag rule-body when jurisdictions diverge.
6. At session end, write session results:
   - If `study-plan.yaml` exists: append to `session_history` per the schema in the `study-plan` skill.
   - If not: write to `~/.claude/plugins/config/claude-for-legal/law-student/session-history.yaml`.
7. Report:
   - Score: X/N (percentage)
   - Missed: list with subtopic tags
   - Weak subtopics this session
   - Pattern vs. prior sessions on this subject (if history has 2+ prior)
   - What the plan now recommends next

© anthropics · Apache-2.0 · anthropics/claude-for-legal

Similar skills

All skills for Agent setup: memory, context, orchestration →

From the same set118

Project sponsors

Companies that keep this analytics open
Gold sponsor

Impressions, clicks and code copies — open statistics. We keep the numbers public for the same reason NeBlask itself is public. Become a sponsor