| name | research-start-lawdroidai |
| title | /research-start |
| description | Research roadmap for a legal issue — statutory starting points, case law areas, agency guidance, secondary sources (NCLC, Shriver Center, practice manuals), search terms for free research tools (CourtListener, Descrybe, Free Law Project) and paid ones if the office has them. Leads and frameworks, NOT authoritative citations; staff verify and develop everything. Use when a staffer asks where to start researching, wants a research roadmap for an issue, or needs gaps identified in existing research. |
| author | lawdroidAI |
| author_url | https://github.com/lawdroidAI/legal-aid-plugin/tree/main/skills/research-start |
| license | Apache-2.0 |
| version | 0.1.0 |
| execution_mode | open |
| jurisdiction | us |
| practice | general |
| language | en |
/research-start
- Load
~/.claude/plugins/config/claude-for-legal/legal-aid/CLAUDE.md → jurisdiction(s), practice areas, funder restrictions.
- Read the practice-area guide at
guides/<practice-area>.md if one exists and the issue belongs to that area.
- Use the workflow below.
- Frame the issue specifically — jurisdiction, procedural posture, what's at stake. Vague framing produces vague roadmaps.
- Build the roadmap: statutory starting points (unverified), case law areas (not specific cases), agency guidance, secondary-source pointers, search terms.
- If the staffer has existing research uploaded: synthesize and identify gaps.
- Output with prominent "leads not authorities" header. Everything is a starting point the staffer verifies.
/legal-aid:research-start "habitability defense to nonpayment eviction in NC"
/legal-aid:research-start "SSI overpayment waiver — without fault standard"
/legal-aid:research-start "U-visa supplement B requirements when no LEA cooperation"
Research Start: Roadmap, Not Research
Purpose
Legal research is essential to civil legal aid practice. But the initial phase — figuring out what to research, finding the right statute, understanding the framework — is often the most time-consuming and least valuable part. A staff attorney can spend a morning finding the starting point before doing any actual research.
This skill produces the starting point: statutory candidates, case law areas to investigate, agency guidance to find, secondary-source pointers (poverty law manuals, practice guides, treatise areas), and search terms for both free tools (CourtListener, Descrybe, Free Law Project) and paid ones if the office has access. None of it is verified. None of it is authoritative. All of it is a lead for the staff attorney to run down.
This is both an ethical safeguard and a practical one. Citations from memory are how malpractice happens. Citations from a roadmap, verified by the staffer, are how good lawyering happens.
Load context
~/.claude/plugins/config/claude-for-legal/legal-aid/CLAUDE.md → jurisdiction(s), practice areas, funder restrictions that might bear on the research scope (e.g., LSC class-action restrictions if the research is heading toward systemic remedies).
The practice-area guide at guides/<practice-area>.md if one exists → office-specific research conventions, judges' preferences, treatises the office subscribes to.
If the office's CMS or document store contains prior research on similar issues (a brief bank, an issue file), reference it — research compounds across matters in the same practice area.
Workflow
