Introducing immigration-legal: An Employment-Based Immigration Practice Toolkit for Partner Attorneys

I just contributed to the Claude for Legal open-source workflow ecosystem with a focus on structured immigration legal workflows.

The idea is simple. Legal AI should not be just a chatbot.

For high-stakes immigration categories like EB-1A, EB-2 NIW, and O-1A, the real value comes from workflow structure, evidence organization, attorney review boundaries, and explainable preparation.

This contribution focuses on how immigration legal workflows should be designed:

  • Each skill should carry its own reasoning scaffold.

  • Shared guardrails should support the workflow, not rescue it.

  • Important claims should be tagged with provenance.

  • Draft outputs should remain attorney-reviewable.

And the system should never confuse preparation support with legal advice or guaranteed outcomes.

This is also the philosophy behind Meritocrat.us.

Applicants prepare. Attorneys decide.

AI should help organize the friction, surface the gaps, and make the case picture clearer before legal strategy begins.

Open-source legal workflows can help the legal tech community move from generic prompting toward structured, reviewable, and responsible AI-assisted legal work.

Here is how the core spine of the workflow is structured.

Phase 1: Practice profile, merit, and evidence

Phase 1 focuses on building a reusable practice profile and giving attorneys a structured path from raw material to legal preparation.

Cold-start interview
Command: /immigration-legal:cold-start-interview

• Captures firm profile, visa mix, and risk posture.

• Records how conservative or aggressive the firm wants to be on each criterion.

• Accounts for service center considerations, comparable evidence posture, writing style, and escalation rules.

The plugin writes this practice profile to:

~/.claude/plugins/config/claude-for-legal/immigration-legal/CLAUDE.md

From that point on, every immigration-legal skill reads from the same profile automatically, so attorneys do not have to keep re-explaining the firm’s standards in each matter.

Merit evaluation
Command: /immigration-legal:merit-evaluation

• Evaluates EB-1A, EB-2 NIW, O-1A, and O-1B matters criterion by criterion.

• Produces a GREEN, YELLOW, or RED conclusion per criterion, grounded in the practice profile.

• Enforces an interpretability rule: every conclusion must cite specific evidence and explain the reasoning behind the classification, rather than hiding behind a black-box score.

Evidence organizer
Command: /immigration-legal:evidence-organizer

• Maps each exhibit to the relevant regulatory criteria and flags coverage gaps.

• Surfaces duplication, under-explained evidence, and unaligned materials that do not yet fit a petition strategy.

• Produces a clearer, criterion-organized preparation layer for attorney review, instead of a jumble of PDFs and emails.

This phase builds the spine of the workflow: the shared profile, structured merit analysis, and a mapped evidence set that make later strategy and drafting meaningful.

From strategy to post-filing: later phases

Once the spine is in place, the plugin turns to structured strategy, drafting support, post-filing work, and day-to-day operations. Many of these skills are currently shipping as explicit stubs: they load, describe their intended behavior, and then hand control back to the attorney until the next build cycle replaces them with full implementations.

Phase 2: Strategy and drafting

• Intake
  Command: /immigration-legal:intake
  Designed for structured new-client intake and high-level cross-visa triage.

• Strategy memo
  Command: /immigration-legal:strategy-memo
  Helps attorneys think through which visa category to pursue, which criteria to lead with, and what evidence development plan is needed before filing.

• Petition letter draft
  Command: /immigration-legal:petition-letter-draft
  Builds an IRAC-style scaffold, criterion by criterion, with rule, evidence inventory, and prompts for attorney analysis.
  The point is not to auto-file petitions, but to accelerate structured drafting while preserving human judgment.

Phase 3: Post-filing support

• RFE response
  Command: /immigration-legal:rfe-response
  Analyzes RFEs or NOIDs, maps them to evidence gaps, and scaffolds a response plan.

• Client update
  Command: /immigration-legal:client-update
  Supports routine client communication in clear language, without diluting the attorney’s professional role.

Phase 4: Operational workflows

• Matter workspace
  Command: /immigration-legal:matter-workspace
  Handles multi-client file management (new, list, switch, close) and stores workspaces at:
  ~/.claude/plugins/config/claude-for-legal/immigration-legal/matters/

• Customize
  Command: /immigration-legal:customize
  Lets firms update a specific section of the practice profile without re-running the entire cold-start interview, so risk posture and house style can evolve over time.

Alongside these skills, the plugin ships with regulatory reference files that act as citation anchors for the merit-evaluation logic:

• eb1a-criteria.md for 8 CFR 204.5(h)(3) EB-1A extraordinary ability criteria.

• eb2niw-prongs.md for the three Dhanasar prongs under EB-2 NIW.

• o1-criteria.md for O-1A and O-1B criteria under 8 CFR 214.2(o)(3).

These files are references, not legal advice. Attorneys still carry responsibility for checking current USCIS guidance and exercising professional judgment.