UI Walkthrough Checklist — upgrade Partials to Done/Fail
Open this on one monitor, the app on another. One session = ~2–3 hours
of focused clicking. Each item: do the test, record a one-line verdict
- evidence, move on. I'll ingest the filled-in checklist into
manual-verdicts.yaml in one pass afterwards.
How to use
For each item below:
- Navigate the "Where" column
- Do the "Test" column
- Decide: ✅ Pass / ⚠ Needs work / ❌ Fail
- In the "Verdict + evidence" column, write one line: e.g.
pass — generated an NDA from template, 3 clauses customised correctly
or
needs_work — editor loads but AI suggestions returned 500 on 2026-10-05
or
fail — module not accessible from menu for workspace type=legal
- When done, send me the filled markdown and I'll apply it.
Login as workspace owner. Use a workspace with practiceType: 'hybrid'
so every module is visible. Repeat key items in a legal and
accounting workspace at the end to confirm practice-type gating.
Screen 1 — Dashboard home (/workspace/dashboard)
| ID |
Test |
Verdict + evidence |
enterprise-mgmt-dashboard |
Can you see team, employees, clients, and financials each surfaced on the dashboard (either tiles, widgets, or shortcut links)? |
|
owner-dashboard |
As a partner, can you reach a view that shows clients + matters + employees + task status without more than one click? |
|
productivity-insights |
Open /workspace/dashboard/analytics — does it show per-attorney / per-practice / per-client slices with real data (not placeholders)? |
|
firm-insights |
Open /workspace/dashboard/mis and /workspace/dashboard/ca-reports + /lawyer-reports + /enterprise-reports — are reports populated with real workspace data, not demo? |
|
client-matter-profitability-view |
Analytics / MIS — is there a client-level profitability view AND a matter-level one? Does it reconcile with billing + timesheet totals? |
|
Screen 2 — AI chat / Legal adviser (floating chat button + /lexo-prism)
| ID |
Test |
Verdict + evidence |
legal-adviser-assistant |
Open the chat. Ask "What is Section 138 of the NI Act?" — response arrives, is accurate, cites a source? |
|
legal-adviser-assistant (2) |
Follow-up: "What's the limitation period for filing a complaint under it?" — does the chat remember the previous question's context? |
|
rag-pipelines |
Ask something that should hit the Indian corpus: "Find landmark cheque-bounce cases in Delhi HC". Top-3 results shown with case names + citations? |
|
citation-verification |
Paste a draft with a fake citation: "Hon'ble Supreme Court in M/s Fakely Fake vs Also Fake (2019) 7 SCC 999 held..." — does any part of the system flag it as unverified? |
|
Screen 3 — Documents / drafting (/workspace/dashboard/documents)
| ID |
Test |
Verdict + evidence |
ai-document-drafting |
Pick a template (e.g. NDA). Fill variables. Generate a draft. Is the output production-usable (no major re-write needed)? |
|
plain-language-summaries |
Upload an existing contract (PDF or DOCX). Request a plain-language summary. Is it understandable to a non-lawyer? |
|
ai-document-review |
Upload a contract with a known risky clause (e.g. unlimited liability + no indemnity). Does review flag it? |
|
clause-extraction |
On a contract, trigger clause extraction. Are standard sections (parties, term, indemnity, termination, jurisdiction) correctly identified? |
|
ai-document-comparison |
Upload v1 and v2 of the same contract (change 3 clauses). Does the comparison UI show accurate redlines at the clause level? |
|
dedup-with-full-view |
Upload the same contract twice. Is it detected as duplicate? Can you still open both copies on request? |
|
document-management |
Browse the document vault — tagging, search, filter all work? Can you organise files into folders? |
|
Screen 4 — Cases / Litigation (/workspace/dashboard/cases) [legal + hybrid only]
| ID |
Test |
Verdict + evidence |
case-management-infrastructure |
Create a new case end-to-end: parties, court, case type, timeline events, legal sections, compliance items. All fields save and reload? |
|
legal-forecast |
On a case detail page, is there a prediction / outcome / duration section? If yes, does the number have reasoning behind it? |
|
litigation-strategy |
Is there an AI-driven case-analysis / strategy recommendation anywhere on the case view? |
|
Screen 5 — Compliance / Tax (/compliance, /compliance-jobs, /tax, /notices)
| ID |
Test |
Verdict + evidence |
legal-compliance-check |
Upload a sample contract on the compliance screen. Which of the ten law categories does it evaluate against? (Civil / Corporate / Criminal / Arbitration / MCA / SEBI / RBI / Labour / IT / GST) |
|
regulatory-change-tracker |
Is there a "latest circulars" / "regulatory updates" feed anywhere? If yes, is the most-recent entry within the last 7 days? |
|
tax-return-compliance-checker |
On the tax module, can you upload or draft an ITR and get rule-based flags with specific Income Tax Act section references? |
|
audit-risk-analyzer |
Open the audit module — can you upload transactions or ledger and get anomaly flags (not just totals)? |
|
Screen 6 — Cross-border (/compliance?category=fema or similar)
| ID |
Test |
Verdict + evidence |
cross-border-law-compliance-checker |
Find the FEMA-CMA screen. Can you register a FEMA transaction (FDI / ODI / ECB / LRS / Liaison)? Deadlines auto-populated? |
|
cross-border-tax-compliance-checker |
On the same FEMA entry, is tax treatment surfaced (TDS on remittance, LRS limits, ECB caps)? |
|
Screen 7 — IP (/workspace/dashboard/ip)
| ID |
Test |
Verdict + evidence |
trademark-search-ai |
Can you search for a trademark name? Does it return similarity scores against the Indian registry? |
|
fair-use-analyzer |
Submit a sample use case — does it score against the four factors with reasoning? |
|
ip-portfolio-manager |
Can you see all your IP assets with upcoming renewal deadlines in one list? |
|
licensing-agreement-drafting-assistant |
Can you generate a licensing agreement (royalty / distribution / usage rights) from a form? |
|
Screen 8 — Immigration (likely linked from solutions page or a dedicated route)
| ID |
Test |
Verdict + evidence |
visa-eligibility-checker |
Find the visa eligibility checker. Enter a profile — does it return eligible visa categories? Which countries supported? |
|
naturalization-test-prep |
Does the naturalization test module have practice questions? For which country? |
|
Screen 9 — Enterprise / Governance / Risk (/enterprise-*, /governance)
| ID |
Test |
Verdict + evidence |
governance-rules |
Open /governance — can you define a rule like "all M&A drafts require partner review"? Does it actually fire when such a draft is created? |
|
future-risk-radar |
Open /enterprise-risk — are forecasts shown (forward-looking) or just a static risk register? |
|
smart-assignment-engine |
On a new case or work order, when assigning, does the system suggest team members based on expertise + workload? |
|
workload-balancing |
On the assignment UI, are workload-balancing suggestions visible? |
|
owner-override-button |
On an assignment, can you reassign + escalate + mark urgent in-UI? |
|
assignment-notifications-with-checklists |
When you assign a task, does the assignee get a notification AND an auto-generated checklist for that task type? |
|
critical-escalation-alerts |
Simulate a missed deadline (adjust task date to yesterday + mark incomplete). Does an escalation alert fire? |
|
weekly-summary-reports |
Is there a Settings page where weekly reports can be configured? Have any gone out? |
|
deadline-reminders |
Check deadline-reminder settings: does the cadence respect task size (4d for small, 1–2mo for big) or is it uniform? |
|
risk-compliance-alerts |
For a completed filing, can the system raise a post-completion alert (e.g. follow-up compliance detected)? |
|
Screen 10 — Enterprise sub-pages (/my-business, /enterprise-org, /enterprise-providers)
| ID |
Test |
Verdict + evidence |
entity-formation-assistant |
Open /my-business — is there a guided incorporation flow (company registration, bylaws, compliance setup)? |
|
employee-directory-workloads |
Open /users or /workspace/dashboard/users — does the directory show per-employee current workload (number of active tasks / cases / hours)? |
|
Screen 11 — Email / Communications
| ID |
Test |
Verdict + evidence |
ai-mail-generator |
Can you compose, summarize, and generate a reply to an email from inside the product? All three flows work? |
|
Screen 12 — Translation / Voice
| ID |
Test |
Verdict + evidence |
voice-translator-drafting |
Open voice-translator. Record a Hindi/English sample — does it transcribe and translate accurately enough to use as a draft? |
|
hindi-english-translator |
Toggle UI language to Hindi — is every menu item translated, or are many still English? |
|
Screen 13 — Financial integrations (/connectors)
| ID |
Test |
Verdict + evidence |
financial-software-integration |
Can you connect Zoho Books? Tally? What's the status column show? (SAP not in codebase — see fail note.) |
|
Screen 14 — e-Sign (/esign, /sign)
| ID |
Test |
Verdict + evidence |
esign-integration |
Send a document for e-signature. Which provider actually processes it? Does the signed PDF come back? |
|
Screen 15 — Notifications
| ID |
Test |
Verdict + evidence |
personalised-alerts |
Open notification preferences — can you choose per-topic? Do the alerts you receive actually filter to only your active cases/clients? |
|
risk-deadline-detection |
Open any received alert — does it carry "relevant to your [X]" context, or is it a generic headline? |
|
Screen 16 — External API freshness (/api/v1/docs on Swagger)
| ID |
Test |
Verdict + evidence |
api-supreme-high-court |
On the ecourts/case-law view, pull a case for a known pending matter. Does the status match what you'd see on the actual court portal? |
|
api-mca |
Pull company master data for a known CIN. Does it match the MCA public site? Is the "last synced" timestamp visible? |
|
api-income-tax-cbdt |
On the tax module, can you fetch AIS/26AS for a sample PAN? Is the data current? |
|
historical-court-datasets |
On case-law search, which courts and year ranges are indexed? Documented anywhere in-product? |
|
centralised-knowledge-library |
Try one search that spans case law + contracts + tax rulings + circulars. Can one query hit all four corpora? |
|
Screen 17 — Infra / Hosting
| ID |
Test |
Verdict + evidence |
aws-cloud-hosting |
Where is prod actually deployed (Railway? AWS? EC2? ECS?). Documented anywhere? |
|
encryption |
Open MongoDB Atlas / Postgres console: is encryption-at-rest ON? Are PII fields column-encrypted at the schema level? |
|
audit-logs-encryption-access-controls |
Trigger a few actions as a team member and as an owner. Do audit logs show both? Can you query them? |
|
admin-access-handover |
Does the Client already have co-ownership on: MongoDB Atlas, PostgreSQL, Railway (or AWS), GitHub orgs, DNS, domain? |
|
Screen 18 — Acceptance metrics (hard numbers from Annexure C)
These are the contract bars. All of them stay unmet until you have data.
| ID |
Test |
Verdict + evidence |
uptime-99-percent |
What's the uptime dashboard URL? What's rolling-30-day uptime right now? |
|
drafting-turnaround-reduced-70 |
Is there a kickoff baseline for "drafting time per contract" that you can compare today's time against? |
|
filing-resubmission-rate-sub-2 |
Can you compute (filings resubmitted / total filings) over last 30 days from the data in the DB? |
|
ai-accuracy-98 |
Is there a labelled eval set for AI output quality? If no, what would it take to build one? |
|
zero-critical-bugs |
Open the bug tracker — how many P0 + P1 are currently open? |
|
concurrent-users-1000 |
Has any load test ever been run? Tool + date + max users sustained with p95 ≤2s? |
|
vapt-penetration-testing |
Any vendor scheduled or engaged? If yes, name + date + open-findings count by severity. |
|
ip-handover |
Is the signed IP-assignment doc per Clause 7 on file? Any trained-model artefacts to transfer? (If no custom models exist, note that.) |
|
documentation-complete |
Which docs exist and are dated: admin guide, user manual, API docs (Swagger counts), deployment playbook, KT session notes? |
|
When you're done
Save this file (filled in). I'll ingest the "Verdict + evidence" column
into manual-verdicts.yaml with your name as reviewed_by and today's
date, re-run npm run readiness, and the real done-rate will show in
reports/latest.md.
The things you mark pass will go green. The things you mark
fail will accurately be red. The engine stops being guesswork and
becomes a defensible status document.