Claude Legal Clinic Plugin
Run Anthropic's legal-clinic plugin for law school clinics doing pro-bono work. 14 skills covering professor setup + student onboarding + case work + supervision + semester handoff. Built for access to justice. ABA Opinion 512 framework. CLINIC credential.
Anthropic shipped a law school clinic plugin on May 12, 2026 — and it’s built for access to justice
The legal-clinic plugin covers law school clinical legal education workflows: client intake, drafting (asylum apps, eviction answers, protective orders, demand letters), IRAC memo scaffolding, research roadmaps, deadline tracking, supervisor review, client communications, and semester handoffs. Built around a clear mission: supercharge access to justice through AI-enabled clinical education.
The problem: clinics are structurally capacity-constrained. A supervising professor manages 5-10 students. Each student carries a handful of cases while juggling classes. Students turn over every semester. Administrative tasks consume hours that could go to client advising. Result: long waitlists, limited caseloads, people who give up waiting.
This plugin cuts the time cost of everything around the lawyering, so the same students and professor serve meaningfully more clients — and students spend more time on the analysis and strategy that make clinical education worthwhile.
This course walks the 14 active skills: cold-start-interview, build-guide, ramp, client-intake, draft, memo, research-start, status, client-letter, supervisor-review-queue, deadlines, client-comms-log, semester-handoff, customize. You’ll learn: two-tier audience design (professor + student); ABA Formal Opinion 512 framework; three supervision models (formal review-queue / ad-hoc / lighter-touch); pedagogy postures (assist/guide/teach); semester turnover via ramp + semester-handoff; cross-practice-area integration (immigration, housing, family, criminal defense, civil rights, consumer protection).
You’ll come out with a CLAUDE.md profile at ~/.claude/plugins/config/claude-for-legal/legal-clinic/CLAUDE.md, a sample end-to-end case workflow, and a credential (CLINIC-XXXXXX) documenting the work. The plugin costs nothing beyond your Claude Pro subscription. The operating discipline this course teaches is the difference between AI assistance that adds capacity and AI assistance that protects pedagogical value.
What You'll Learn
- Setup clinic via `cold-start-interview` + per-practice-area `build-guide` (professor work)
- Run `ramp` for student semester onboarding + practice exercises
- Operate `client-intake` + `draft` + `memo` + `research-start` for case work scaffolding
- Use `status` + `client-letter` + `client-comms-log` for case communications
- Track `deadlines` + use `supervisor-review-queue` for supervision; complete `semester-handoff`
After This Course, You Can
What You'll Build
Course Syllabus
Prerequisites
- Active law school clinical professor OR law student working in a clinic
- Familiarity with clinical legal education + supervision framework
- Claude Cowork or Claude Code installed; ability to install plugins from the Anthropic marketplace
Who Is This For?
- Clinical legal professors at law schools
- Law students working in clinics (immigration, housing, family, criminal defense, civil rights)
- Clinical fellows + adjunct clinical professors
- Pro bono coordinators at law schools
- Legal aid attorneys collaborating with clinics
- Bar associations supporting clinical legal education
Frequently Asked Questions
ABA Formal Opinion 512 (2024) — how does the plugin operate within this framework?
Opinion 512 says lawyers may use generative AI but must: maintain competence in the technology, maintain confidentiality, supervise outputs, communicate with clients about AI use where appropriate, and verify before relying. The plugin operates within this: confidence markers signal verification needs; supervisor review structures hold; cold-start asks about supervision style; client-letter doesn't auto-send. Plugin scaffolds the work in ways compatible with attorney supervision.
How does the plugin handle the semester turnover problem?
Each semester, new students need to learn procedures, tools, practice areas. `ramp` skill is interactive onboarding — reads the clinic handbook professor uploaded at setup; teaches it; provides low-stakes practice exercises (fake intake, practice draft, research roadmap) before real cases. `semester-handoff` produces per-case handoff memos for next cohort. Together: reduces ramp time from weeks to days; preserves continuity across semesters.
Three supervision models — what's the trade-off?
Formal review queue: every student output queued for professor review before client communication. Tight oversight; clear audit trail. Ad-hoc: professor reviews as needed; less structured; faster student work. Lighter-touch: standard safeguard labels on every output; professor supervises through existing clinic structure (case rounds, one-on-ones). Choose during cold-start based on clinic size + practice area sensitivity + professor preference. Switchable later.
Does the plugin replace clinical pedagogy?
No. The plugin scaffolds the work so students can do more lawyering per semester. Pedagogy is the professor's domain. The plugin's design: students do analysis + strategy (the educationally valuable parts); plugin handles intake + first drafts + research roadmap + status summaries (the time-consuming non-educational parts). Net effect: same students serve more clients without sacrificing the learning experience.
How does the plugin handle confidentiality with multiple students?
Each case has isolated client-comms-log; per-case file separation; access controls via clinic IT setup. Student turnover requires explicit access transfer via semester-handoff. Plugin's design treats each case as confidential matter; doesn't allow cross-case context without explicit configuration. Account tier (Team vs Enterprise) for confidentiality is an open question per clinic IT/ethics review.
