For solo and small immigration firms
The RFE is in. The clock is running. The evidence is everywhere.
Every request matched to a folder of scans, the exhibit index renumbered, the cover letter rewritten, while new consultations wait and the deadline doesn’t move. On a flat-fee matter, every extra hour comes straight out of your margin.
The Small Firm AI Kit for Immigration: one Claude Skill with 17 ready-made playbooks.
- RFE and NOID response outlines
- Evidence indexes and cover letters
- Declaration outlines
- One-time payment
- 30-day money-back guarantee
- Works on free Claude
RFE evidence map and response outline (excerpt)
Item 1, commingled finances (RFE p. 2). On file: Bank 1 joint statements, January and February 2026 (Exh. C-1, C-2). Gap: only two months on file for a longer marriage. Client request: additional months of joint statements and a joint tax return, if one was filed.
Item 2, shared residence (RFE p. 2). On file: joint lease signed September 2025 (Exh. D-1). Suggested support not yet in the folder: utility bills or mail addressed to both spouses at the same address.
Checked before you see it
- Checked: All 3 RFE items mapped, none skipped
- Checked: Every exhibit reference matches a file in your upload
- Flagged: 2 evidence gaps flagged, with a client request list
If this sounds like your week, keep reading
- The RFE buries its requests in dense pages, and you’re matching each one to a folder of scans.
- You renumber the exhibit index and rewrite the cover letter by hand because one document changed.
- Your declaration interview notes run to pages, and turning them into an outline that still sounds like your client takes another evening.
- A client asks again where the case stands, and you write the update at night, between filings.
Where the hours go
Roughly 12 hours a week go to the same four jobs
The same week with Claude and these playbooks: about 5 hours back
- Today
- With The Small Firm AI Kit for Immigration
RFE and NOID responses
Today: 4 h/wkWith The Small Firm AI Kit for Immigration: 2.5 h/wkExhibit indexes and cover letters
Today: 3 h/wkWith The Small Firm AI Kit for Immigration: 1.5 h/wkClient updates and time entries
Today: 3 h/wkWith The Small Firm AI Kit for Immigration: 1.5 h/wkDeclaration interviews and outlines
Today: 2 h/wkWith The Small Firm AI Kit for Immigration: 1.5 h/wk
Today: ≈12 h a week, ≈576 h a year on this work.
Back in your week, our estimate
≈5 h
Example starting value: $200. Move it to your own number.
- Pro pays for itself after
- ≈24 min
- Time back, each month
- ≈$4,000
How we estimated these numbers
Only part of a small-firm lawyer’s week ever gets billed. A big share of the rest goes to paperwork that repeats on every file. This is a conservative picture for a small immigration practice with one RFE or NOID every two weeks and a filing going out most weeks. If your caseload is heavier, so are these numbers.
- RFE and NOID responses, 4 h/week: Our estimate: about 8 hours to map the requests to the evidence and build the response, one RFE or NOID every two weeks. No published figure found.
- Exhibit indexes and cover letters, 3 h/week: Our estimate: about 3 hours per filing, one filing a week. No published figure found.
- Client updates and time entries, 3 h/week: Our estimate: about 36 minutes a working day on status emails and time-entry narratives. No published figure found.
- Declaration interviews and outlines, 2 h/week: Our estimate: about 4 hours per declaration, interview included, one every two weeks. No published figure found.
These are our estimates, not measured results. No published study times these immigration tasks, so we started from the conservative hours above and assumed every task still needs you: the legal argument, the client interview, checking every exhibit reference, clearing every [VERIFY] and signing off. So we cut each task by half at most, and declarations by only a quarter, because the interview stays yours. For comparison, in the 8am 2026 Legal Industry Report 38% of lawyers who use AI say it saves them 1–5 hours a week and 24% say 6 or more.
Week: A standard 40-hour week (our basis). Clio’s 2025 Legal Trends for Solo & Small Firms puts utilization at 27% for solos and 32% for small firms: at those rates only about 11–13 of the 40 hours get billed.. Yearly figures use 48 working weeks, monthly ones 4.
One Skill · 17 playbooks inside
Describe the case. See the draft it builds.
One Skill in Claude: you describe the case in plain words, its router opens the right playbook, and back comes a draft with every exhibit referenced. These are excerpts of what comes back, built on fictional, anonymized facts.
Every request pulled apart and matched to evidence, with nothing skipped.
You ask
“I-130 RFE for Petitioner A and Beneficiary B attached, plus the case folder. Map every request to the evidence we have and what’s missing.”
You get
Each RFE item against the exhibits on file, and the gaps still to fill.
Playbook: rfe-noid-decoder.md · in every plan
immigration-lawyers/
├─ SKILL.md
the router: reads your request, opens the right playbook
├─ your profile, saved at setup
└─ playbooks/
├─ citation-guard.md
├─ confidentiality-preflight.md
├─ engagement-letter-ai-clause.md
└─ … 14 more
See all 17 playbooks
In every planBasic · $39
- Every citation flagged before anyone else sees it
citation-guard.md - Client data anonymized before any work starts
confidentiality-preflight.md - Client updates your clients actually read
client-update-writer.md - Time entries clients don’t push back on
billing-narrative-cleaner.md - Every RFE item mapped to evidence and a response outline
rfe-noid-decoder.md - An exhibit index and cover letter from your case folder
evidence-index-cover-letter.md - Declarations in your client’s own words
declaration-interview-outline.md
Added in ProPro · $79
- An engagement letter with an AI clause, ready in minutes
engagement-letter-ai-clause.md - A document checklist for every visa and benefit type
client-document-checklist.md - Support letters that don’t sound like a template
support-letter-scaffold.md - Form data pulled from ID documents into a table you verify
form-data-extractor.md - A clear structure for translated evidence
translation-summary-scaffold.md - A clean case summary straight from intake
immigration-intake-case-summary.md - Client updates for every stage of the case
case-stage-updates.md - Interview prep built from your client’s own file
interview-prep-pack.md - A first-draft RFE response, argument by argument
rfe-response-draft.md - Every receipt, deadline and next step in one table
case-deadline-tracker.md
Why not just ChatGPT?
Why not just paste the RFE into ChatGPT?
Around 70% of solo and small-firm lawyers already use AI (8am 2026 Legal Industry Report; Clio 2026), but only 9% have an AI policy they actually enforce (8am). A plain chatbot is fast, but it writes every client in the same voice and doesn’t know which exhibit backs which claim. In immigration, that gap shows up in tribunals and credibility findings.
Generic AI
- UK and Munir v SSHD [2026] UKUT 81 (IAC): one adviser cited a case that doesn’t exist and uploaded client letters to an online AI tool. The Upper Tribunal said that puts the information in the public domain and loses privilege. In the second case a trainee drafted judicial review grounds with false authorities, and the supervising solicitor was referred to the SRA.
- A UK immigration solicitor pasted emails containing Home Office decisions into ChatGPT to “improve” them. Judicial review forms now require a statement of truth that the authorities cited actually exist (Law Society of Ireland Gazette, 2026).
- Matter of R-K-K- (BIA 2015): significant similarity between declarations in unrelated cases can support an adverse credibility finding. Formulaic AI phrasing creates exactly that similarity, and one firm’s blog reports more asylum clients drafting declarations with ChatGPT.
- Generic hardship language in I-601 and I-601A letters draws RFEs, and AI-drafted National Interest Waiver petitions have shown fabricated citations and inflated metrics (Herman Legal Group).
The Small Firm AI Kit for Immigration
Why this works where a plain chatbot didn’t
- 1
Anonymized before anything starts
Confidentiality Pre-Flight runs first on every task: it offers anonymized mode, swaps names for Petitioner A and Beneficiary B, and reminds you to use a business plan that doesn’t train on your data.
- 2
Every request pinned to an exhibit
RFE/NOID Decoder ties each request to the page of the notice and the exhibit on file, and lists what is missing. Evidence Index + Cover Letter flags anything unmatched, and statements without a source are marked [NO SOURCE].
- 3
No invented cases, no ghost-writing, your signature last
Citation Guard never generates an authority and marks every citation [VERIFY]. Declaration Interview Outline quotes the client verbatim. Every output ends with a verification block and a line for your sign-off.
How it works
Set up once. Then just ask.
No coding, no prompts to copy. You install it the way you would add a document to Claude.
- 1
Download the ZIP
Right after checkout you get one ZIP, ready to download on the page: a single Claude Skill for your practice with every playbook in your plan inside. No account, no login, no subscription.
- 2
Upload it to Claude
Add it once in Claude’s settings. After that you just ask for the work: the Skill reads your request and opens the right playbook, so you never pick a file yourself. It runs on every Claude plan, including Free with code execution turned on. For real client files, use a business plan that doesn’t train on your data.
- 3
Answer the 5-minute Firm Profile interview
Country, agency or court, regulator, house style, signature and how you reference evidence (exhibit, tab or page). Saved once and read by every playbook, so no draft starts in the wrong forum.
Compare your options
What the other options cost
These are the real alternatives a solo or small immigration firm has today, at the prices reported for them. None of them is wrong. They are priced for a different firm.
- Case-management seat with AI (Docketwise)$69–119 per user per monthA practice-management system whose assistant handles correspondence and notes; reviewers say it doesn’t generate full petition packets. Billed every month, per user.
- Legal AI subscription (CoCounsel, Lexis+ AI)CoCounsel ~$220–639 a month; Lexis+ AI ~$128–494 a month on top of Lexis+ (third-party pricing, 2026)Built for legal research and billed every month per user. It doesn’t map an RFE to your exhibits.
- Your own hoursYour hourly rate, request by requestThe same mapping, index and cover letter rebuilt for every filing, against the deadline.
- A generic chatbotAbout $20 a monthFormulaic phrasing is the similarity Matter of R-K-K- warns about, and in Munir the Upper Tribunal said uploading client letters to open AI tools loses privilege.
- Claude’s official legal pluginsFree to install, paid Claude plans onlyNo immigration plugin and nothing that drafts an engagement letter AI clause.
- The Small Firm AI Kit for Immigration$39 or $79, once$39 or $79, paid once: about what one case-management seat costs for a single month, and it never renews.
Pick your plan
One payment. Yours to keep.
About one month of a single case-management seat, with no renewal after it. One payment, yours to keep.
Pro costs, once
$79
Basic: $39 once
It pays for itself after
≈24 min
of time won back, at $200 for an attorney hour
Get back just 1 hour a month
$200
a month, against a one-time $79. Our estimate is ≈5 h a week.
Break-even = plan price ÷ the hourly value set in “What those hours are worth” above.
Basic
The core kit plus the three immigration essentials.
$39one-time
7 ready-made playbooks
Try it on real work. Not for you? Full refund within 30 days.
- 4 core playbooks: Citation Guard, Confidentiality Pre-Flight, Client Update Writer, Billing Narrative Cleaner
- RFE/NOID Decoder: every request mapped to evidence and a response outline
- Evidence Index + Cover Letter from your case folder
- Declaration Interview Outline that keeps the client’s own words
- Firm Profile interview: set your forum and house style once
- Verification block at the end of every output
Pro
The full immigration workflow, from intake to decision.
$79one-time
All 17 playbooks
Only $40 more than Basic: +10 playbooks and the Pro bonuses below.
Try it on real work. Not for you? Full refund within 30 days.
- Everything in Basic
- Client Document Checklist for each visa or benefit type
- Support Letter Scaffold with generic-language and inflated-metric warnings
- Form Data Extractor: one ID document at a time, every field to verify
- Translation Summary Scaffold
- Immigration Intake → Case Summary
- Client updates for every stage of the case
- Engagement Letter + AI Clause
- Interview Prep Pack for your client
- RFE Response Draft, argument by argument
- Case Deadline Tracker
- Word templates for your cover letter, exhibit index and RFE response
- Firm Style Guide: your house style, applied by every playbook
- Sample outputs from every playbook, to compare against your own
- 12 months of fixes and adjustments to the included playbooks
30-day money-back guarantee
Run the kit on a real, anonymized file. If it doesn’t earn its place in your practice within 30 days, email us and we refund every cent. No conditions.
- One-time payment, no subscription
- Works on free Claude
No Claude account yet?
And if I don’t have Claude?
Claude is Anthropic’s AI assistant. The free plan is enough to run The Small Firm AI Kit for Immigration: creating an account takes a minute, and you don’t need a credit card.
- Free plan works, no subscription needed to start
- Account ready immediately
- Paid plans only add more usage, for when you want it
- 1
Create a free account
Go to claude.ai and sign up with your email, Google or Apple account. It’s free and ready right away.
- 2
Turn on one setting
In Settings → Capabilities, switch on “Code execution and file creation”. Skills need it, and it’s free.
- 3
Upload your package
In Customize → Skills, click “+”, choose “Upload a skill” and select the ZIP you received. Then just ask.
Fair questions
What immigration lawyers ask before buying
Is it ethical to use AI on client matters?
In the US, ABA Formal Opinion 512 doesn’t forbid it. It asks you to understand the tool, verify its output, protect confidentiality (informed consent may be needed), communicate with the client, supervise staff and charge reasonable fees. In the UK the SRA’s concerns are false citations and client data in public tools. The kit is built around both: an AI clause for your engagement letter, anonymize-first by default and a human-verification block on every document. It is aligned with the guidance; it doesn’t certify compliance anywhere.
Is my client data safe?
Every task starts with Confidentiality Pre-Flight, which offers anonymized mode and swaps names for placeholders. Use Claude’s Free plan only to try the kit with anonymized or fictional data; for real client files, use a business plan that doesn’t train on your data. The Upper Tribunal’s warning in Munir is the reason this step comes first.
Won’t AI-assisted declarations hurt my client’s credibility?
They can, if AI writes them. That is why the Declaration Interview Outline playbook doesn’t. It guides the interview and organizes your client’s own answers, quoting them verbatim with a reference to where each came from, and it flags phrasing that repeats across declarations, the similarity Matter of R-K-K- warns about.
Will it invent cases?
Citation Guard never generates an authority. It extracts every citation in a draft, marks it [VERIFY] and tells you whether it appears in the documents you uploaded. We don’t promise zero errors. We promise every citation is flagged for you to check before you file.
Doesn’t Anthropic already do this?
Claude’s official legal plugins, released in May 2026, focus on commercial, corporate, litigation and similar work. None is built specifically for immigration practice, and none drafts an engagement letter AI clause. These playbooks fill that gap and run on Claude’s Free plan, while the plugins need a paid plan. We are independent and not affiliated with Anthropic.
Built for your rules
Built for the forum you practise in
Choose your practice setting in the Firm Profile and the playbooks switch vocabulary, structure and the rules they remind you of.
Your setup
USCIS benefits
Family, employment and waiver filings: RFEs, NOIDs, cover letters and support letters.
- RFE/NOID Decoder maps each request to an evidence checklist and a response outline
- Support Letter Scaffold warns on generic hardship language (I-601/I-601A) and inflated metrics (NIW) (Pro)
- Form Data Extractor turns a passport, green card, EAD or I-94 into a field table you verify (Pro)
- Client document checklists for each visa or benefit type (Pro)
Guardrails
Verification isn’t a disclaimer. It’s the product.
Every output ends with the same verification block: key statements, the source of each, the items marked [VERIFY], the human checks this document requires and a line for your signature. We never promise zero errors. We promise every citation is flagged for you to check.
Sample verification block
From: RFE evidence map and response outline (excerpt)
- Checked: All 3 RFE items mapped, none skipped
- Checked: Every exhibit reference matches a file in your upload
- Flagged: 2 evidence gaps flagged, with a client request list
- Verify: Draft citations: 2 found in the RFE, 1 NOT FOUND in your upload. Verify before filing
- 01
Citation Guard
No authority is ever generated. Every citation is marked [VERIFY] and checked against your uploads: found in your documents, or NOT FOUND, verify before filing.
- 02
Declarations stay in the client’s words
Interview, outline, verbatim quotes tied to the client’s answers. No stock emotional sentences, and a check that flags phrasing repeated across declarations.
- 03
An exhibit reference on every fact
Every statement in a response outline or cover letter points to its exhibit or page. Anything without a source is marked [NO SOURCE].
- 04
Every extracted field marked to verify
Data pulled from passports, I-94s and EADs comes back as a table with every field flagged for you to check against the original.
- 05
Anonymize first
Before any task, anonymized mode is offered and names become Petitioner A and Beneficiary B. Consumer plans are for trying the kit, not for client files.
- 06
Jurisdiction before anything else
USCIS, immigration court, Home Office or Upper Tribunal? Asked once in the Firm Profile and read by every playbook.
- 07
Deadlines stay with you
The playbooks read the deadline from the notice and mark it [VERIFY]. They never calendar it for you; your docketing system and your own check do that.
- 08
Disclosure wording when rules ask for it
Suggested AI-use paragraphs for engagement letters and filings, plus a statement-of-truth reminder for UK judicial review, marked as drafts to adapt to your local rule.
FAQ
Everything else
Didn’t find your answer? Email us:hello@summonskills.com
What exactly is in the ZIP?
One Claude Skill for your practice. Inside: a router file (SKILL.md) that reads each request and opens the right playbook, one Markdown playbook per job (7 in Basic, 17 in Pro), each with its instructions, templates and verification block, and the Firm Profile interview every playbook reads. Pro also includes the Firm Style Guide inside the Skill, plus a bonus pack: Word templates for the cover letter, exhibit index and RFE response and sample outputs from every playbook. 12 months of fixes and adjustments to the included playbooks are included (new playbooks for future rule changes are not part of that).
Does it work on Claude’s Free plan?
Yes. The Skill runs on every Claude plan, including Free, as long as code execution is turned on. Free is fine for trying the kit with anonymized or fictional facts. For real client files, use a business plan that doesn’t train on your data.
I use ChatGPT. Will this work there?
No. The kit is built as a Claude Skill, and ChatGPT has no equivalent way to install it once and have it open the right playbook for each request. Claude’s free plan is enough to try it, an account takes a minute, and you can keep ChatGPT for everything else.
Does it cover the UK as well as the US?
Yes. It is built first for US practice before USCIS and the immigration courts. Set the UK in your Firm Profile and the playbooks work in Home Office and Upper Tribunal terms, with the Munir and SRA safeguards switched on.
Will it track my RFE deadline?
It reads them, it doesn’t calendar them. In Pro, the Case Deadline Tracker pulls the dates from your receipt notices and correspondence into one table, each marked for you to check against the notice. It never works out a deadline or puts anything in a calendar: your docketing system and your own check stay the safeguard.
Does it replace my case-management software?
No, it works alongside it. Export the notice, documents or notes from whatever system you use and upload them to Claude. No connector needed.
What if it isn’t right for my practice?
Email us within 30 days and we refund the full price. No conditions.
Is this legal advice?
No. The kit is for licensed attorneys, solicitors and their staff. It drafts and organizes; you review, decide and sign. You remain responsible for every filing and every letter that leaves your office.
Your next RFE, mapped to the evidence before you start writing.
Every request linked to an exhibit, every gap listed, every declaration in your client’s own words, every citation flagged and your sign-off last. About 5 hours a week back, by our estimate. $39 or $79, paid once. Use it on a real, anonymized file, and if it doesn’t earn its place within 30 days, we refund every cent. Not legal advice; for licensed lawyers.
One payment, yours to keep · 30-day money-back guarantee